#note: in Australia so way less problem with healthcare costs
Explore tagged Tumblr posts
Text
Me: man it’s so annoying that my dad won’t go to the dr when something is obviously wrong, it sucks that he waits until he is really ill
Also me: I do not need to go to Dr about this, I’ll just cope, it can’t be that bad
14 notes
·
View notes
Link
In an example to the rest of the scientific community and an effort to wake up people—particularly policymakers—worldwide, 17 scientists penned a comprehensive assessment of the current state of the planet and what the future could hold due to biodiversity loss, climate disruption, human consumption, and population growth.
"We aim to provide leaders with a realistic 'cold shower' of the state of the planet that is essential for planning to avoid a ghastly future." — 17 scientists
"Ours is not a call to surrender—we aim to provide leaders with a realistic 'cold shower' of the state of the planet that is essential for planning to avoid a ghastly future," according to the perspective paper, co-authored by experts across Australia, Mexico, and the United States, and published in the journal Frontiers in Conservation Science.
Co-author Paul R. Ehrlich of Stanford University's Center for Conservation Biology—who has raised alarm about overpopulation for decades—told Common Dreams his colleagues "are all scared" about what's to come.
"Scientists have to learn to be communicators," said Ehrlich, citing James Hansen's warning about the consequences of "scientific reticence." Hansen, a professor at Columbia University's Earth Institute and former director of the NASA Goddard Institute for Space Studies, testified to Congress about the climate crisis in 1988.
Ehrlich was straightforward about how "extremely dangerous things are" now and the necessity of a "World War II-type mobilization" to prevent predictions detailed in the paper: "a ghastly future of mass extinction, declining health, and climate-disruption upheavals (including looming massive migrations), and resource conflicts."
"What we are saying might not be popular, and indeed is frightening. But we need to be candid, accurate, and honest if humanity is to understand the enormity of the challenges we face in creating a sustainable future," said co-author Daniel T. Blumstein of the Institute of the Environment and Sustainability at the University of California, Los Angeles, in a statement about the paper.
"By scientists' telling it like it is, we hope to empower politicians to work to represent their citizen, not corporate, constituents," he said in an email to Common Dreams.
The paper, Ehrlich and Blumstein pointed out, comes in the midst of the coronavirus pandemic—which, according to Johns Hopkins University, has killed nearly two million people. Over the past year, the Covid-19 crisis has provoked calls for humanity to end its destruction of the natural world to prevent future public health catastrophes.
"We're all seeing the shocks to our global systems now from Covid and the rise of authoritarian leaders," Blumstein said. "Because our current ways of life are ecologically unsustainable (we're living in an ecological Ponzi scheme), we fully anticipate more—and more deadly—pandemics in the future. We expect civil unrest, wars, and famines. We are all shaken by the likelihood of the collapse of civilization as we know it."
The new warning from scientists, Blumstein noted, cites over 150 other papers "documenting the diverse and shocking decline in biodiversity and planetary 'health' and their consequences." Among the cited sources is a World Wide Fund for Nature (WWF) report that in September revealed an "average 68% decrease in population sizes of mammals, birds, amphibians, reptiles, and fish between 1970 and 2016."
"In the midst of a global pandemic, it is now more important than ever to take unprecedented and coordinated global action to halt and start to reverse the loss of biodiversity and wildlife populations across the globe by the end of the decade, and protect our future health and livelihoods," WWF International director general Marco Lambertini said at the time.
The co-authors—including William J. Ripple of Oregon State University, who last year led thousands of scientists in declaring a climate emergency and earlier this month led a call for "a massive-scale mobilization to address the climate crisis"—echoed Lambertini's message while also underscoring the importance of increasing awareness about what's actually needed.
"Humanity is causing a rapid loss of biodiversity and, with it, Earth's ability to support complex life. But the mainstream is having difficulty grasping the magnitude of this loss, despite the steady erosion of the fabric of human civilization," the paper says.
"In fact, the scale of the threats to the biosphere and all its lifeforms is so great that it is difficult to grasp for even well-informed experts," said lead author Corey Bradshaw of Australia's Flinders University in a statement. "The problem is compounded by ignorance and short-term self-interest, with the pursuit of wealth and political interests stymieing the action that is crucial for survival."
The paper explains that "while suggested solutions abound, the current scale of their implementation does not match the relentless progression of biodiversity loss and other existential threats tied to the continuous expansion of the human enterprise." According to its authors, "That we are already on the path of a sixth major extinction is now scientifically undeniable."
"With such a rapid, catastrophic loss of biodiversity, the ecosystem services it provides have also declined," the paper explains. Consequences include "reduced carbon sequestration, reduced pollination, soil degradation, poorer water and air quality, more frequent and intense flooding and fires, and compromised human health."
Highlighting estimates that the human population will near 10 billion by 2050, the scientists lay out how "large population size and continued growth are implicated in many societal problems," from food insecurity, soil degradation, biodiversity loss, and an increased chance of pandemics, to crowding, joblessness, deteriorating infrastructure, and bad governance.
"Recycling, using less plastic, eating less meat, taking public transportation, and flying less, while all important, will simply not create the rapid change we need now to save much of the Earth's biodiversity and our lives." —Daniel Blumstein, UCLAThe paper also details the planetary impacts of dirty energy and carbon-intensive food production, and says that "while climate change demands a full exit from fossil fuel use well before 2050, pressures on the biosphere are likely to mount prior to decarbonization as humanity brings energy alternatives online."
A section on failed international goals declares that "stopping biodiversity loss is nowhere close to the top of any country's priorities, trailing far behind other concerns such as employment, healthcare, economic growth, or currency stability."
"The dangerous effects of climate change are much more evident to people than those of biodiversity loss, but society is still finding it difficult to deal with them effectively," the scientists note, while decrying "utterly inadequate" efforts by governments to even try to meet the targets of the landmark Paris climate agreement.
They further decry the recent rise of right-wing, anti-environment agendas in countries including Australia, Brazil, and the United States—which recently denied President Donald Trump a second term. Ehrlich expressed hope that President-elect Joe Biden will work to deliver on the climate promises he made as a candidate.
Biden's vow to rejoin the Paris agreement "is positive news," but "it is a minuscule gesture given the scale of the challenge," Ehrlich said in a statement.
The president-elect "is moving in the right direction," Ehrlich told Common Dreams, pointing to the selection of former Secretary of State John Kerry as his climate envoy. However, "the Paris goals are increasingly looking inadequate," and "Biden's political opportunities to do anything major may be greatly constrained," he added.
Blumstein stressed that "recycling, using less plastic, eating less meat, taking public transportation, and flying less, while all important, will simply not create the rapid change we need now to save much of the Earth's biodiversity and our lives."
According to Blumstein, "We need rapid political change."
He urged voters to elect leaders who will end fossil fuel use as well as "eliminate perpetual economic growth and properly price externalities so that the environmental costs are built into the price of a product." He also emphasized the importance of access to education and reproductive control, and the need to rein in corporate lobbying and enact campaign finance reform so politicians serve citizens' needs.
"Ultimately," Blumstein added, "we must focus on making equity and well-being society's goals—not the constant accumulation of more junk."
In their paper, the UCLA scientist and his 16 co-authors "contend that only a realistic appreciation of the colossal challenges facing the international community might allow it to chart a less-ravaged future."
It is "incumbent on experts in any discipline that deals with the future of the biosphere and human well-being to eschew reticence, avoid sugar-coating the overwhelming challenges, ahead and 'tell it like it is,'" they conclude. "Anything else is misleading at best, or negligent and potentially lethal for the human enterprise at worst."
1 note
·
View note
Text
This Too Shall Pass
As investors, we take a multi-year view. While we do not want to minimize the devastation to life caused by the coronavirus nor its near term impact to the global economy, our role is to look over this valley, as this too shall pass, and take advantage of the inefficiencies created in all of the financial markets.
Global stock markets and commodity prices are falling while bond prices are soaring. The dollar has regained its strength and safe haven status along with the Japanese yen. Opportunities are being created for the patient, long-term investor, but go slowly as uncertainty is the word of the day and markets hate uncertainty.
So far there are approximately 12,000 reported cases of coronavirus worldwide and nearly 260 deaths globally, all in China. If history is any guide, the rate of change in the number of people getting the virus and dying from it should begin to slow in a matter of weeks. The Chinese government is moving rapidly to contain the virus, corporations are shutting down their operations asking employees to stay at home, governments are restricting travel from and to China, and the best of the best in healthcare are working together to control, mitigate, and ultimately find a cure for the virus.
Let’s put the coronavirus in perspective. The flu has infected over 19 million people while killing over 10,000 so far in the 2019-2020 flu season in the U.S alone. No one talks about it as a national disaster and a detriment to growth as we feel that we have it under control. Do we? The health authorities, when approving a vaccine each year for the flu season, rarely get it right. Does the flu epidemic alter the long growth potential of the U.S? No. Will the coronavirus alter the long-term growth potential of China? Not really! So why not look over the valley and take advantage of these near-term inefficiencies in the financial marketplaces? It is not easy for sure, but the best opportunities are created when others are uncertain and panic. We maintain a long-term view: buy when value exists and sell when fully valued. We always challenge our core beliefs and remain open to change, if/when warranted. We fully recognize that we live in a VUCA (volatile, uncertain, complex, and ambiguous) environment but invest for long term gains rather than trade which is a losing strategy over time.
Our base case remains that the coronavirus and the flu will both be contained before the end of the first quarter, 2020. There is no doubt that near term global economic growth will be penalized big time as growth in China in the first quarter could easily be halved from previous expectations and multinational earnings will clearly be impacted too. But will that alter long-term prospects for both China and these corporations. Not at all!
Chinese monetary authorities mentioned yesterday that they will provide all the capital needed to deal with the economic blow from the coronavirus to support their financial markets. We fully expect the government to announce a massive additional fiscal stimulus plan to jump start the Chinese economy once the virus is contained. In fact, all of the global monetary authorities will keep the money spigot wide open in 2020. Our Fed, which met this past week, confirmed its easy monetary policy not expecting the funds rate to increase until inflation reaches 2% for a sustained period which won’t happen anytime soon. On the other hand, Fed Chairman Powell did mention that the Fed may begin to wind down its $60 billion per month expansion in its balance sheet sometime this spring but that remains to be seen during these questionable times. The Fed will continue, no matter, to expand its balance sheet in the future but at reduced levels to maintain ample reserves in the system so the repo problem does not happen again.
Investment Conclusion
The engines for accelerating global growth are primed and ready to go once the world gets its arms around the coronavirus, trade deals kick in, and monetary stimulus is in effect, reducing the cost of capital for businesses and consumers alike. Add to all of this, the benefits of major fiscal stimulus in China, Japan, and the U.S. If anything, the virus may cause both monetary authorities and governments to do even more than is currently on their plates to re-ignite their economies.
We expect the global economy, as well as that of the U.S, to improve as we move through the year. For example, the U.S economy alone will be hit by around 0.5% annualized during the first two quarters of 2020 due to Boeing’s problems bringing the Max back online. We are hopeful that the FAA will approve it by mid-summer with production beginning even sooner but at much lower levels than existed in 2019. On the other hand, all of the trade deals finally concluded by the U.S. with China, Mexico, Canada, Japan and others will ramp up, too, as we move through the year, potentially adding well over 0.7% to annualized growth in the latter half of 2020. Net net, we would not be surprised to see first half GNP growth slightly less than 2% with the final two quarters running at 3% or above. In addition, as we wrote above, China’s growth will accelerate rapidly with added monetary and fiscal stimulus once the coronavirus is controlled such that their economy will recover and sustain growth at around 6% annualized once again as the year progresses.
Just the acceleration of these two economies alone would be sufficient to boost global growth as we move through the year, but we also expect the emerging markets, Japan, Australia, India and others to move into gear as the year progresses. By the way, India’s government just slashed taxes and increased fiscal spending to boost growth back above 5% in 2020 from beneath 4% last year. You can see why we remain optimistic that global growth will end the year on a high note, after a sluggish start, which is not the consensus at this time, which we like.
After having reduced our economically sensitive stocks over the last few weeks as discussed in previous blogs, we are looking to add them back over the next few weeks as they are beginning to be priced, once again, at recession valuations. Each of the companies that we would buy have strong managements with winning strategic long term plans, have exceptionally strong free cash flow after investing in their businesses, have dividend yields well over 3% and growing and are buying back tons of stock out of free cash flow reducing the count by over 3% per year. Energy is not part of this group.
Before we conclude this blog, we must discuss two areas of uncertainty that has existed in the marketplace over the last few months. First, it now is evident that the Senate will vote next week to exonerate Trump. The impeachment proceeding will finally come to an end. Second, Brexit is a done deal as of Friday night at 11:00 PM. The transition phase will last for a year with Britain as a nonvoting member of the Eurozone but be able to trade freely within the EU. If the two sides cannot reach a deal on their future relationship by December 31st, business will be conducted on World Trade Organization terms and border checks will be imposed where now none exist. We would not be surprised to see Trump and Johnson make a major trade deal between them before either one makes one with the Eurozone.
In conclusion, we believe that the major issues facing investors today are transitory and that growth will resume as we move through the year. We are also factoring in that all of the monetary authorities are unusually accommodative providing far more capital to the system than is needed by the real economy thus finding its way into risk assets just like last year. And you know how the financial markets reacted last year! Remember that most all interest rates have retreated to near all-time lows, spreads remain tight and major bank capital/liquidity ratios are at all time highs. Clearly the stock market multiple should exceed 20 times projected earnings.
But the major difference today than last year is that we are on the cusp of accelerating global growth with all the stars aligned once the coronavirus is controlled. This too shall pass. This has happened time and again when epidemics spread in parts of the world. Now is the time to average into stocks when uncertainty is high, valuations are low, dividend yields exceed the 10-year treasury yield, and few are optimistic about the future. We expect to look back in a few months after the virus is controlled and ask why we did not do even more buying as others panicked. We recognize that the Chinese markets could have a blood bath when they reopen next week but we expect the government to do all in its power to mitigate the decline. We will continue to average into those stocks that we want to own long term as prices weaken.
Our major area of emphasis remain technology, including the semis. Look at Microsoft’s numbers. It is our second largest position. We also own some financials who are earning more each year, generating tremendous free cash flow, pay large and growing dividends and are shrinking their capitalizations big time. This has all been done with a relatively flat yield curve. Can you imagine their earnings once the yield curve normalizes?
Global capital goods and industrials remain a major focus, too, as their volumes and earnings grow despite sluggish global growth. They are generating over 100% free cash flow per year that is in the billions being used to enhance shareholder returns. Also, we own the low-cost industrial commodity companies that are also generating billions of free cash flow with large well supported dividend yields over 5% and large buybacks in place. We expect shortages in copper in 2021. Housing related retailers is another area that we favor as there is a shortage of low cost housing in this country. Finally, we own many special situations where their intrinsic value is at least 50% greater than current valuations. Each one has exceptionally strong balance sheets with huge free cash flow, too and yield over 3%. We own no bonds and are flat the dollar.
Our weekly investment webinar will be held Monday at 8:30 am EST. You can join by entering https://zoom.us/j/9179217852 into your browser. Feel free to send questions in advance to [email protected].
Remember to review all the facts; pause, reflect and consider mindset shifts; turn off the pundits/experts; look at your asset mix with risk controls; listen to as many earnings call as possible; do independent research and …
Invest Accordingly!
Bill Ehrman
Paix et Prospérité LLC
1 note
·
View note
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://venabeahan.tumblr.com
0 notes
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://wittooth.tumblr.com/
0 notes
Text
Online MBA in Australia - Saves Your Precious Time
To make the choice of the best business schools, you will get plenty of opportunities and also options. Online MBA in Australia is a part of it. This management degree is the ideal way of entering the corporate world and step over the ladder of success. It is perhaps the most coveted professional degrees of this time. There are many accredited online business administration schools where you can seek admission.
There are many colleges and universities in Australia offering Online MBA in Australia Programs. But the main thing of concern is to carry on the education seamlessly, since the cost of a management program is expensive. So unless you can arrange for the financer then it will be almost impossible on your part to continue with the education.
An MBA is a master’s degree program in the administration of the business. It will give you the opportunity to know the details of how to run business properly. An Online MBA is that type of product, where you will get management education through the online route. An MBA degree will help to move your career ahead of times and the online route will help you to get the degree at the most inexpensive way. The most important thing of getting an online degree is that you will get it at your own convenient time. But consider getting admission in an accredited program. The solid business training you will get during your course of study.
Various business schools in Australia are offering Online MBA in Australia degrees. The cost of studying in Australia institutions is much less, when compared with their European or American counterparts. But try to find an accredited college or learning institution, whose degree will not land you in any future problem. This will be most probably in the case of jobs. You will not get the right job, since those colleges are not recognized by the industry bodies.
If you are employed or studying full time for other courses, then an online MBA will help you to get some extra qualification. You will be able to study at your convenient time. You will face no problem, regarding the flexibility of the course. Through a dedicated internet connection and a computer, you will be always ready to access your online course. There may be the option of online classes or taking of notes and study materials. You will also get the option of business training and that may be available also at your choice. Then there are many companies offering you employment opportunities. They will only look at the level of accreditation of any institution. This will help them to find out the appropriate candidate for the job. They will not look at the process of study but the very authenticity of the degree.
Will an Online MBA in Australia Degree Help your Stagnant Career?
Trying to improve work situations and move up in the business world, many people have found that they are being passed over for candidates with better credentials. Economic influences have forced employers to become much more cautious in their hiring selections to make the most of their own constrained resources. When everyone, including yourself, is trying to get more bang for their buck, will an Online MBA in Australia degree help your stagnant career? Not surprisingly, the answer is an unequivocal "Yes."
Access to information through the Internet has changed nearly every aspect of our lives, perhaps the most significant area being furthering education online. The shifting trend continues to reflect the growing number of people opting for online MBA programs to bolster their resumes - whether the result of personal choice, a desire to advance within the ranks of current employment, or in response to economic pressures that continue to impact job markets. As employers have pulled in their horns, the hiring process has nearly become a hybrid process. It only makes sense that the natural rhythm to the reaction from the work force is to turn to the hybrid MBA.
This is an MBA program that offers instruction with more limited access to the on-campus curriculum. Online MBA in Australia programs appeal to those who want the prestige that comes from attending a brand-named school without having to give up their current jobs or make the trip to campus classes every other night of the week. The most attractive advantage to the accelerated MBA degree is that it accommodates modern life, allowing students to maintain their personal and professional commitments and responsibilities. Motivated individuals can take two classes during each session. This is an intensively compacted program that enables the student to achieve their Online MBA in Australia in about half the time and at a fraction of the cost that it takes to accomplish the same thing on campus.
However, the same aspect that enables students to accelerate their learning also tends to keep them from peer networking and career guidance that typical students receive. There is little faculty face-time, and personal interaction with other students is limited by comparison to more conventional programs. What online learning provides includes emails, forums, a variety of academic tools, chat rooms and other discussion boards, all of which are making their way into the mainstream methods of conducting business today.
The ease and convenience of taking courses online is the primary surveyed reason most people opt for either Online MBA in Australia or part time MBA programs. Some people do better at different hours of the day, and these programs allow the student to choose when they want to study. You can learn at your own pace, which means slow learners can review materials that may be difficult, and arrangements can be made with the instructor for private tutoring.
Schools are beginning to favor these online, hybrid and part time MBA programs to help boost revenue, especially right now as the downturned economy has had a negative impact on endowments and alumni donations. It makes sense that these Online MBA in Australia programs don't cost the campus as much, as fewer resources and services are committed to students. Accelerated MBA degrees are obtainable for Healthcare Administration, Finance, Accounting, Entrepreneurship & Managing Innovation, International Business, Marketing, Non-profit Management and Real Estate Studies.
Applications continue to increase as more students find that Online MBA in Australia programs work extremely well to complete the required degree needed to advance in their positions, often in overseas appointments. As hybrid students are not typically going head-to-head for jobs that their full-term (2 year) counterparts are vying for, they are simply adding this degree to boost their resume without having to indicate that it was achieved from a hybrid MBA program. This is happening as those offering the jobs have become much more comfortable with students obtaining degrees online.
For getting more information visit here VIT - Victorian Institute of Technology.
14/123 Queen St, Melbourne VIC 3000, Australia
1300 17 17 55 (or) [email protected]
#MBA Melbourne#MBA Australia#MBA Sydney#Online MBA Australia#Online MBA Sydney#Online MBA Melbourne#Master of Business Administration in Australia#Master of Business Administration in Melbourne#Master of Business Administration in Sydney
0 notes
Text
Claire Interview
Who are you and what business did you start?
My name is Claire I am CEO and co-founder of Kalix. My background is as a dietitian.
Kalix is a fully featured telehealth, practice management, electronic documentation and electronic medical record for private practice dietitians and other healthcare professionals. It is a completely web-based solution or software as a service (SaaS).
Telehealth technology supports the provision of health services electronically (or virtually) facilitating long distance patient and clinician contact. The telehealth features Kalix offers includes HIPAA compliant virtual meeting (video conferencing), secure client messaging, online appointment scheduling, electronic forms and agreements, online payments and electronic food and exercise tracking.
Practice Management Software are solutions that manages the day-to-day operations of private healthcare practices. The have a range of features including; appointment management and scheduling, appointment notification, client management and billing (insurance billing and online payments).
Electronic medical records are electronic information files containing information about individual client medical history and health.
Electronic documentation systems assist with the systematic documentation of client medical information, clinician-patient interactions and correspondences electronically.
Being web-based, Kalix is available online world-wide. We follow a subscription model, users subscribe to Kalix and are charged either on a monthly or yearly basis for ongoing access. Subscriptions and billing are automatically through the website (using Stripe).
We currently have over 1000 paying users from 19 countries. The majority of customers are from the United States of America- (90% of users). Other countries include Australia, Canada, New Zealand, Israel, Arab Emirates, Switzerland and Trinidad to name a few.
What’s your backstory and how did you come up with the idea?
I trained and previously worked as a dietitian. I graduated with a Master of Science (Nutrition & Dietetics) with Distinction through the University of Wollongong in Australia. Working in both the public health sector (acute, subacute and outpatient) as well as private practice, I was very lucky to experience a diverse work history.
The idea for Kalix arose while I was working as the sole dietitian in an orthopedic surgery early intervention team. The position was funded by a special government grant, so I was under a lot of pressure to measure and evaluate the effectiveness of my professional practice.
With a heavy workload and stuck using a documentation/tracking system built for medical staff not dietitians, I needed to get creative. While investigating how to measure and evaluate professional practice, I came across the Nutrition Care Process Terminology (NCPT). NCPT is standardized language developed by the Academy of Nutrition and Dietetics. It describes the specifics of what a dietitian does. Because NCPT is standardized and covers all the data that a dietitian collects during initial and follow-up assessments, I surmised that if I was to write all my patient notes using NCPT, a software system should be able to track changes in the variables associated with NCPT terms. Tracking the changes in these variables would be an easy, sensitive and efficient way of evaluating my professional practice without having to spend extra time measuring, recording and analyzing data. The statistical analysis could be built into patient documentation and hence save me a lot of time.
The problem was finding a software system that uses NCPT in this way. I needed a program that supports quick electronic documentation using NCPT terms, tracks of changes in patient data over time and correlates changes in the variables. There was not a system like this available on the market.
Co-founder Felix Jorkowski was looking for a new product to undertake. After 4 months of initial development work in September 2012, we first launched Kalix at the International Congress of Dietetics (ICD 2012) held in Sydney. By attending ICD 2012 were able to expose and receive feedback about Kalix from dietitians from around the world. We invited key stakeholders to an “interest group”. We also paid for flyers to be included attendee’s welcome bag. Kalix was also made available to use free online by anyone.
Describe the process of designing, prototyping, and manufacturing the product.
As described above, Kalix was initially launched as a documentation system only, not a complete electronic medical record and practice management solution. The decision for this change in direction came after analyzing the feedback we received at ICD 2012. While we initially identified public hospital dietitians as our target market, there was a major issue we kept bumping into. Each hospital already had their own electronic medical record system and the dietetic departments required their integration with the Kalix system. They also not willing to pay the cost of this integration. Unfortunately, this was not a viable option for us at the time, so we decided to explore different options.
There was a lot of ongoing interest for Kalix from dietitians working in private practice, especially those in the US. While dietitians working in the hospital system already had their own EMR system, private practice dietitians on the other hand, were still most commonly using paper-based systems or stuck with ill-fitting EMRs and practice management solutions designed for medical staff and other professions. There was no practice management solution/ electronic medical record designed specifically for dietitians. It became increasingly evident that dietitians working in private sector should be the target market we pursue.
Kalix was relaunched in May 2013 as a complete practice management solution. Our initial aim was to build the minimal viable product make it available online, receive customer feedback and continue adding to it. In August 2013 we also received $40,000 seed funding for product commercialization and the development for the US market. Australian private practice dietitians are a very small market and we needed to expand to other countries to be successful. There are over 68 000 Dietitians in the US, compared to less than 3000 in Australia.
The major hurdle to sell in the USA was achieving HIPAA compliance. Health Information Portability and Accountability Act (HIPAA) is the data privacy and security legislation that describes how electronic health information is transferred, received, handled, or shared. With a few modifications to Kalix and lots of assistance from legal specialists, we achieved HIPAA compliance in October 2013. That month we also exhibited Kalix at the biggest trade show for nutrition professionals in the USA. After receiving much useful feedback, we completed some fine tuning and started selling Kalix in the US during January 2014. The United States quickly became our primary market and the company was relocated to Los Angeles in August 2015.
Describe the process of launching the online store/business.
Our goal for Kalix has always been to build a minimal viable product and make it available online for use and purchase. This allows us to receive customer feedback right away, so that we can continue to fine tune and release new features based on popular request.
Kalix’s co-founder, Felix is a software developer so there were no costs associated with site development and making it available online. Kalix was part of the Microsoft BizSpark program (now discontinued) which gave us free website hosting and access to other services for three years. We paid to work with various graphic designers for site design and branding including freelances and a personal friend that provided us with discounted rates. Being a little bit arty myself I taught myself to use Illustrator and Photoshop to complete some of the design work.
To create product awareness we attended several tradeshows in both the United States and Australia. Through product demos, the handing out flyers and the collection email addresses for marketing lists we were able to generate strong interest. We were able to get Kalix listed on as a suggested resource on a number of external websites utilized by our target market. This was very important for establishing credibility. We achieved over 900 sign ups during the initial launch period.
Since launch, what has worked to attract and retain customers?
Because Kalix follows a subscription model, retaining customers is just as important (or even more so) than attracting new ones.
Regular communication with customers to encourage engagement and the frequent release of new features, helps keep existing customers invested in Kalix. Adapting Kalix’s development schedule based on the changing market requires has also been beneficial. We continuously update and improve existing features to improve workflows customers have trouble with.
How are you doing today and what does the future look like?
Kalix has experienced ongoing growth since its initial launch. Our net revenue has increased by 44% over the past 12 months. Our subscriber churn rate continue to be quite low, currently sitting at 2.3%. The life time value of each of our customers is $1,140.
We are currently in the process releasing a new subscription level at a higher cost point. The new features released as part of the telehealth component of Kalix are very popular, but costly. Increasing the price is required to maintain our profitability. We are currently working on how to manage this price increase without losing customers.
We are also working on market expansion target other non-medical healthcare profession e.g. therapists, social workers, speech pathologist. No modifications are required to Kalix to target these professions.
Through starting the business, have you learned anything particularly helpful or advantageous?
Marketing is just as important as the product develop side of things. You can have the best product in the world but is no one knows about it; your effects are wasted. The saying “If you build it, they will come” is just not true any longer.
Learning from your mistakes is a must. Customers generally are understanding when things can go wrong. As long as we are responsive and take actions to prevent the issue from recurring you can bounce back.
What platform/tools do you use for your business?
Intercom – we use Intercom for all of our customer support, onboarding and marketing requirements.
Status Page – provides our customers with real-time and historical data on system performance downtime and scheduled maintenance.
Stripe – online payment processing platform. Our customers’ subscriptions and billing are automated through the though the Kalix website using Stripe. Stripe also have analytics to track growth, customer retention etc.
Xero – our accounting software.
SocialPilot – social media scheduling tool, to create and schedule social media post ahead of time across multiple platforms.
Canva and Place it – for the generation of social media images.
Advice for other entrepreneurs who want to get started or are just starting out?
Complete your research, check for other similar products that are already available, is there room for one more? If it is something brand new, will there be a large enough demand? Talk to your target market and receive feedback early and often. Find out if (and what) they willing to pay for your product. Do not be afraid to change direction in product development if needed.
Are you looking to hire for certain positions right now?
We looking to hire staff in customer support roles within the next 2 – 3 months. Please feel free to email [email protected].
Where can we go to learn more about you?
Website: https://www.kalixhealth.com/
Facebook: https://www.facebook.com/KalixHealth
Twitter: https://twitter.com/kalixemr
Blog: https://blog.kalixhealth.com
LinkedIn: https://www.linkedin.com/in/clairejnichols1/
The post Claire Interview appeared first on Facebook Advertising Agency | Facebook Marketing Company.
from Facebook Advertising Agency | Facebook Marketing Company https://voymedia.com/claire-interview/
0 notes
Text
As I’ve always suspected, Health Care = Communism + Frappuccinos
By MATTHEW HOLT
Happy 15th birthday THCB! Yes, 15 years ago today this little blog opened for business and changed my life (and at least impacted a few others). Later this week we are going to celebrate and tell you a bit more about what the next 15 years (really?) of THCB might look like. But for now, I’m rerunning a few of my favorite pieces from the mid-2000s, the golden age of blogging. Today I present “Health Care = Communism + Frappuccinos”, one of my favorites about the relationship between government and private sector originally published here on Jan7, 2005. And like the Medicare one from last week. it sure holds true today. Matthew Holt
Those of you who think I’m an unreconstructed commie will correctly suspect that I’ve always discussed Marxism in my health care talks. You’d be amazed at how many audiences of hospital administrators in the mid-west know nothing about the integral essentials of Marx’s theory of history. And I really enjoy bring the light to them, especially when I manage to reference Mongolia 1919, managed care and Communism in the same bullet point.
While I’ve always been very proud of that one (err.. maybe you have to be there, but you could always hire me to come tell it!), even if I am jesting, there’s a really loose use of the concept of Marxism in this 2005 piece (reprinted in 2009) called A Prescription for Marxism in Foreign Policy from (apparently) libertarian-leaning Harvard professor Kenneth Rogoff. He opens with this little nugget:
“Karl Marx may have suffered a second death at the end of the last century, but look for a spirited comeback in this one. The next great battle between socialism and capitalism will be waged over human health and life expectancy. As rich countries grow richer, and as healthcare technology continues to improve, people will spend ever growing shares of their income on living longer and healthier lives.”
Actually he’s right that there will be a backlash against the (allegedly) market-based capitalism — which has actually been closer to all-out mercantilist booty capitalism — that we’re seen over the last couple of decades. History tends to be reactive and societies go through long periods of reaction to what’s been seen before. In fact the 1980-20?? (10-15?) period of “conservatism” is a reaction to the 1930-1980 period of social corporatism seen in most of the western world. And any period in which the inequality of wealth and income in one society continues to grow at the current rate will eventually invite a reaction–you can ask Louis XVI of France about that.
But when Rogoff is talking about Marxism in health care what he really means is that, because health care by definition will consume more and more of our societal resources, the arguments about the creation and distribution of health care products and services will look more like the arguments seen in the debates about how the government used to allocate resources for “guns versus butter” in the 1950s. These days we are supposed to believe that government blindly accepts letting ��the market” rule, even if for vast sways of the economy the government clearly rules the market, which in turn means that those corporations with political influence set the rules and the budgets (quick now, it begins with an H…). That’s how defense has always been and how pharmaceuticals will increasingly be. Rogoff recognizes the centrality of this argument in his description of what’s wrong with American health care:
“Part of the rise in U.S. healthcare costs stems from the breakdown of the checks and balances that more centralized systems provide. (For example, Americans are several times more likely to receive heart bypass surgery than Canadians, where the procedure is reserved for extreme cases. Yet several studies suggest that patients are no worse off in Canada than in the United States). And even the most fanatical free marketers recognize that healthcare is different from other markets, and that the standard supply-and-demand principles don’t necessarily apply. Consumers have poor information, and there is an obvious case for greater government involvement than in other markets.”
But he then goes on to say that the much greater spending seen here as compared to Canada and the UK creates both a terrible service level (and by implication quality level) and diminishes innovation in health care services. And if all countries squeezed profits in the health sector the way Europe and Canada do, there would be much less global innovation in medical technology.
“Today, the whole world benefits freely from advances in health technology that are driven largely by the allure of the profitable U.S. market. If the United States joins other nations in having more socialized medicine, the current pace of technology improvements might well grind to a halt. Even as the status quo persists, I wonder how content Europeans and Canadians will remain as their healthcare needs become more expensive and diverse. There are already signs of growing dissatisfaction with the quality of all but the most basic services. In Canada, the horrific delays for elective surgery remind one of waiting for a car in the old Soviet bloc. And despite British Chancellor Gordon Brown’s determined efforts to rebuild the country’s scandalously dilapidated public hospital system, anyone who can afford to go elsewhere usually does.”
His conclusion is that because for the sake of social equity government intervention in the system is warranted, the health sector will be a “battleground” between capitalism and socialism through this century. If you get past his mis-use or mis-understanding of the terms “capitalism” and “socialism”, the point he’s making is quite interesting. It does though suffer from a typically Amero-centric bias. Rogoff assumes that the extra spending on health care in America leads to better services and by implication better quality. But that’s an old chestnut. By that measure the higher spending in Canada (11% of GDP) should lead to a better system than in France (9%) or Germany (10%). But in those two nations access to drugs and technology is much greater than in the UK or Canada, and things like waiting times are comparable to the US — in fact in Australia and New Zealand they’re better than they are here. A few years back The Economist said that the Swiss system (again several percentage points cheaper than here) was better than the American on an absolute level. Furthermore recent studies of international care quality suggest that particularly for primary care, the US is results-wise(at best) in the middle of the pack. All of those nations have a heavier proportion of government funding of health care spending than in the US, and all of them spend a whole lot less money. Note that the US government spends more per head (and damn nearly as much as share of GDP) on health care as the whole of the UK.
So that all tells us one thing. We’re paying a lot more for health care here, but it isn’t necessarily getting us better outcomes, innovation or even services. We might though have nicer waiting rooms and we certainly lead the league in surgeons with Porsche 911s. Therefore it’s a stretch to imply that higher private spending leads directly to innovation and better services, particularly if the system is not set up with either government-based or real market-based co-ercive capabilities to promote efficiency and value for money. And lets be real, the US system is set up to provide revenues and profits for providers and suppliers. It’s a bit like saying Tammany Hall provided the best government services because it cost the most, when huge chunks of the money were getting diverted off into corruption.
Furthermore, it’s also a stretch for Rogoff to suggest that by definition government spending creates lower innovation compared to private spending. After all government spending led to the creation of the Internet and biotechnology. Private spending created reality TV. And despite the fact that there is no private spending on defense, well the boys and girls in the US military are no longer riding around on horses pulling gun carts. Somehow innovation and progress seems to find a way to happen even in government sponsored sectors. And if we want to drag real communists into the equation, the reason that we’re not all speaking German is that Hitler lost WWII to a nation that ten years before he invaded was inhabited by peasants. Yup, unpleasant as it might have been, Stalin’s Great Leap Forward in the 1930s was by far the fastest period of economic growth seen in any nation, probably any time…..just in time to save our arses in 1942-4.
I’m not exactly advocating purges, slave labor camps, collectivization and enforced Ten Year Plans as a panacea for the future of health care (although David Brailer keeps going on about his ten year plan). But the overall point is that greater government involvement in spending and regulation of health care doesn’t necessarily mean the disaster in services and innovation that Rogoff suggests. And there are excellent reasons from the “socialist” angle for greater government involvement in health care than we have now.
The first is the fallacy that there can be such a thing as a private health insurance market with free use of underwriting. Social insurance (or universal insurance), in which everyone pays in and everyone receives at least a basic level of benefits is the only way to get around the problem of the uninsured and the uninsurable. It of course means a relative redistribution of income from the healthy and wealthy to the poor and sick, but in fact that can be budget neutral to the healthy and wealthy if the overall price tag is kept down. That though would require a redistribution of income from the health care sector to the rest of society. Such universal insurance is good enough for everyone over-65 in this country and good enough for everyone else in the developed world, but the concept just can’t seem to get the attention of the American public enough to force it past the “special interests” in Congress. And everyone (apart from actuaries and underwriters and some participants in the system) suffers as a result.
The second is the role of government or someone like it as a clearinghouse of information or as a standards-setting body in a market where information access is very lopsided. Health care is very, very complex and someone has to provide decent information (preferably with some regulatory teeth) so that consumers/patients are not at the mercy of providers and suppliers who know far more than they do and in whom most patients still are forced to place their trust blindly. This is the role of the NICE in the UK, and in theory ought to be the role of the FDA here. Adding an economic element to that role by giving information on value for money would probably be derided as socialism by Rogoff’s “capitalists”, but is a rational role for government. And one they are likely to add as spending increases — of course the Brits and Aussies already have done so to some extent, and are linking cost-effective performance to payment.
So overall I don’t think there’s any basis for suggesting that if we have more “socialism” in health care — and by that I’m using Rogoff’s meaning of government spending, regulation and income redistribution — we will necessarily have worse services or lower innovation. Although we may have lower drug prices and a less profitable health care industry. Anyone awake during the last three months of Vioxx breast-beating is becoming painfully aware that expensive “innovation” can be costly for the wrong reasons and actually not be innovative–COX-2s didn’t really do what they were supposed to do (reduce GI problems) but they did cost a lot more than NSAIDs in both money and increased heart disease. But it’s that kind of “innovation” that Rogoff correctly says that Americans are paying more for than anyone else.
However, Rogoff is making a very important point when he discusses the likely trade-offs between basic health care and lifestyle enhancements that will dominate the politics of health care for the next century. We’ve already seen this begin with the medicalization of social afflictions (ugly teeth, small breasts), the medicalization of several “diseases” that aren’t really diseases (impotence, shyness), and the medicalization of old age (osteoporosis, prostate cancer). Now the nano-gurus are discussing the medicalization of death — which will presumably lead to a cure, or at least a delay, for it at a hell of a price.
As more and more health care services become luxury goods, there is a justifiable discussion about what’s a basic necessity and what should someone have to pay for out of their discretionary income. At the moment no-one’s seriously suggesting that your boob job or teeth whitening should be other than an individual expense, or that your cancer treatment is a luxury good to be chosen if your mood and wallet fits. But clearly the middle of that continuum will continue to fill up.
This leads me to what has been called the mocha-Frappuchino problem. I read an article once (that I can’t find anymore) that discussed the increase in productivity of the US workforce since the 1930s. It’s doubled. Which means that we could work half the time and have a 1930s standard of living, or we could work as hard as we do now and have more stuff. The author noted that in the 1930s you couldn’t get a Mocha Frappuchino; so you’ve been spending Wednesday 1pm through Friday afternoon working for your Frappuchino (or similarly frothy goods and services).
We’ve always thought of health care as an “essential”. And eventually even in the US I believe we’ll figure out a way to solve the problem of creating an equitable and sustainable social insurance model for that “essential”. But increasingly, the health care Frappuccino will be paid for and delivered privately, in a separate system. Of course it’s the blurring of those two systems that concerns bleeding heart liberals like me, as that can well lead not as it has done here for the Medicare population, as society giving Frappuccinos to everyone, but instead society deciding to take away essential services from those who can’t afford Frappuccinos.
And that will be the real socialism versus capitalism battle of the next decades.
Matthew Holt is the founder of The Health Care Blog
Article source:The Health Care Blog
0 notes
Text
Can Multiculturalism Work? Part II: Liberal Compassion Unmasked
Why do leftists oppose nationalism? It’s a question at the heart of the problem we’re attempting to grapple with in this series, which in turn grew out of an earlier case for a liberty-oriented nationalism, National Liberty.
A quick recap: last time, I noted that I had made a mistake by attempting to defend a vision of nationalism, rather than first placing the burden of proof on the multiculturalists, specifically those multiculturalists who defend a “post-national” or (in my formulation) “anti-hegemonist” vision of multiculturalism.
(function(w,d,s,i){w.ldAdInit=w.ldAdInit||[];w.ldAdInit.push({slot:10817585113717094,size:[0, 0],id:"ld-7788-6480"});if(!d.getElementById(i)){var j=d.createElement(s),p=d.getElementsByTagName(s)[0];j.async=true;j.src="//cdn2.lockerdomecdn.com/_js/ajs.js";j.id=i;p.parentNode.insertBefore(j,p);}})(window,document,"script","ld-ajs");
We also noted the connection between leftist multiculturalism and the claim of the Left to be the Party of Compassion. In other words, there is a vision on the Left (NAXALT disclaimers notwithstanding) of Omni-Compassionate Welfarist Multiculturalism, a vision in which the famous “liberal compassion” (much of which is bound up with support for a social-democratic welfare state) is wedded to a vision of multiculturalism.
Going forward, let us abridge the unwieldy Omni-Compassionate Welfarist Multiculturalism to Omni-Compassionism, denoting the leftist doctrinal position on compassion which forms the central basis of so much of their political and social thought.
This is the central problem at the heart of this series: how to make sense of this chimerical monster, this fantastical minotaur at the heart of the Labyrinth of left-wing emotion, rhetoric, and theorizing.
In the last article and elsewhere, I have previously noted that leftists only oppose nationalism when done by White people. They’re actually quite supportive of identitarian concerns for every other group, though they generally stop short of full nationalism—and this makes sense, given that they need White, particularly White male, tax dollars to fund their welfare state.
This is the connection between the two streams we’re actually trying to cross in this series: the Left, the Empathy Worm Party, relies on a doctrine of “compassion” to sell people on their ruinous ideas.
But what’s so great about liberal compassion?
Writing for the New York Post, Kyle Smith argues that “liberal compassion is an inane basis for political action:”
“Compassion creates a co-dependency between the empathizers and the empathizees: Liberalism is an ‘alliance of experts and victims,’ in the words of Harvard’s Harvey Mansfield. … “If the empathizees ever disappeared (by, say, having their miseries extinguished) then the empathizers would lose their purpose, the soothing soul balm of caring. “That’s why, no matter how much money gets poured into poverty alleviation ($22 trillion, not including Social Security or Medicare, in the last 50 years), ‘You will never hear the words “Phew! We cured poverty’ “Indeed, official poverty measures simply leave out anti-poverty benefits. That’s like saying you’re carless if you were forced to accept a Chevy paid for by other taxpayers. “So the poverty rate never really changes — it’s been around 14% or so for the last half-century — even as living standards for today’s poor surpass those of yesterday’s middle class.”
For some time, I found the idea of the Left as the Party of Compassion irritating and irksome, chiefly because so many leftists seem to be so self-righteous about it. Last time I linked this article from Justin Rosario, on The Daily Banter, titled: “Why Do Liberals Unfriend People Over Politics? Because Conservatives Are Soulless Monsters.” Here’s the blurb:
“You may not be a neo-Nazi, belong to the KKK, or join white power marches but if you still vote Republican, you've consciously chosen to stand with the absolute worst people in America. Why the f*** would any liberal want to associate with you?”
Honestly, if you’re not laughing at this over a glass of wine or other beverage of choice (as a neo-reactionary, I find wine suitably aristocratic), you’re missing out.
The entire leftist project is one of redistributionism in the name of “compassion.” I don’t dispute that many leftists are genuinely compassionate people who do want to do good works. Even if we set aside the case that conservatives and people of faith are actually more generous than liberals and non-believers (granted, not all of these categories are mutually exclusive), though, I simply don’t find “compassion” to be a substitute for reason.
The real rebuttal to leftist compassion is not to argue that the conservative Christians so many leftists claim to despise may actually have them beat (see links above) in the charitable giving department, a position that is not uncontested. A far more powerful line of critique is to observe that leftists often seem unconcerned with how their elaborate ideas of governmental redistributionism and societal re-engineering will play out in practice.
As author William Voegeli writes for Imprimis of Hillsdale College, in an article titled “The Case Against Liberal Compassion”:
“In fact, however, liberals do not seem all that concerned about whether the machine they’ve built, and want to keep expanding, is running well. For inflation-adjusted, per capita federal welfare state spending to increase by 254 percent from 1977 to 2013, without a correspondingly dramatic reduction in poverty [emph. added], and for liberals to react to this phenomenon by taking the position that our welfare state’s only real defect is that it is insufficiently generous, rather than insufficiently effective, suggests a basic problem. To take a recent, vivid example, the Obama Administration had three-and-a-half years from the signing of the Affordable Care Act to the launch of the healthcare.gov website. It’s hard to reconcile the latter debacle with the image of liberals lying awake at night tormented by the thought the government should be doing more to reduce suffering. A sympathetic columnist, E.J. Dionne, wrote of the website’s crash-and-burn debut, ‘There’s a lesson here that liberals apparently need to learn over and over: Good intentions without proper administration can undermine even the most noble of goals.’ That such an elementary lesson is one liberals need to learn over and over suggests a fundamental defect in liberalism, however—something worse than careless or inept implementation of liberal policies [emph. added]. “That defect, I came to think, can be explained as follows: The problem with liberalism may be that no one knows how to get the government to do the benevolent things liberals want it to do[emph. added]. Or it may be, at least in some cases, that it just isn’t possible for the government to bring about what liberals want it to accomplish.”
Let me be blunt: the problem with Omni-Compassionism is that it is not smart or rational. It is, in practice, often very stupid and very irrational.
It's not that I don’t think we need compassionate people in this society (leftists, consider this your olive branch for this article), but when compassion is touted as the reason for overlooking financial responsibility, there’s a problem. When it is touted as the reason for overlooking unintended, harmful consequences, the problem is rather more severe.
And the fact of the matter is that so many of the policies progressive leftists champion produce unintended consequences, and are often dismal failures.
Minimum wage laws, for example, kill jobs, leaving workers unemployed unemployed or with their hours cut, and encourage companies to replace replace them with robots. Rent control, to give another example, consistently leads to housing shortages and abandoned buildings .
I’ve addressed that sacred cow of leftist policy, “universal healthcare,” before:
“You know things are bad when even the BBC admits, despite some hedging, that the UK’s National Health Service (NHS) is looking at (even) longer waiting times, and is racking up deficits despite a budget of over £130 billion. Even the New York Times is admitting things are getting substantially worse. Meanwhile, the U.S. beats the UK on five-year survival rates for cancers of the breast, prostate, lung, bowel, stomach, liver, and ovaries, not to mention leukemia and ischemic stroke. “Canada? In 2016, median wait times for medically necessary treatments and procedures hit 20 weeks. “Meanwhile, in Sweden, private health insurance is on the rise. It’s quicker—the Swedish public system has terribly long waiting times.”
Making government the monopoly provider of healthcare destroys the entire price-signal function: the market can no longer allow producers and consumers to negotiate prices, so a team of planners must do it more or less arbitrarily.more or less arbitrarily.
If you still think “universal healthcare” sounds like a good deal, staggering taxes and all, consider the system that it replaced:
“During the 19th and early 20th centuries, health care was offered in various ways, including through voluntary mutual-aid associations in Britain, Australia, and the United States. Roderick Long wrote about these fraternal societies, where members could subscribe to various services, including life insurance, disability insurance, and lodge practice… “The average cost of lodge practice for each member was between one and two dollars (a day’s wage) annually, whereas non-members paid the same price for each visit to the doctor [emph. added]. Doctors competed for lodge contracts, which kept costs low. The Canadian experience with lodge practice was similar, and, as in America and Britain, this infuriated the medical establishment.”
It’s all very well and good to say: “I want to help people.” It’s another thing entirely to say: “I’m conscripting you and the rest of society to my vision of how to best help people.” If your actions then say: “I’m completely indifferent to the consequences of the policies I prescribed in the name of compassion,” we have a real problem.
This is the problem we addressed last time: Omni-Compassionism is based on reality-denial. It is a skilled evasion of reality that allows people to feel good about themselves, at the cost of inflicting very real—indeed, as we are learning here, sometimes catastrophic—damage on other people and their societies.
Which is better, the capitalist café or restaurant in Seattle that is willing to employ you for $10.00 an hour and not a penny more, or the leftist government that requires them to pay you $15.00 an hour (50% more, if you’re keeping track), costing you the job as a result? How smart and rational is Omni-Compassionism in this situation?
Which is better, affordable direct primary care run on a no-insurance capitalistic model, or a massively expensive model of government-provided healthcare? Again, understand that Option B is the Omni-Compassionate answer.
Which is better, a markets and enterprise system of producers and consumers that responds to incentives, or a set of idealistic and well-intentioned government policies that consistently produces dreadful and sometimes calamitous consequences? The ‘Omni-Compassionists’ of the Left have certainly made their feelings clear.
Relieving people of the responsibility to be adults and actually take care of themselves, as the Left’s holy welfare state does, produces dreadful personal and social consequences for those people, for the communities they live in, and for the society that hosts their social parasitism.
As Don Feder, writing for the Heritage Foundation, wrote all the way back in 1997:
“In the name of compassion, we've created a welfare state undreamed of in the annals of bureaucratic history. For pity's sake, we've spent over $5 trillion fighting poverty over the past 32 years. All of that compassion has bought us multi-generational welfare families, 80 percent illegitimacy in some inner cities, boys raised in homes without fathers, and rampant crime and addiction.”
Writing in response to the rioting and looting that erupted in many mostly-Black inner-city communities in response to the entirely justified shooting of Michael Brown, Thomas Sowell dismissed the notion that a “legacy of slavery” has anything to do with the social problems that plague these communities:
“We are told that such riots are a result of black poverty and white racism. But in fact — for those who still have some respect for facts — black poverty was far worse, and white racism was far worse, prior to 1960. But violent crime within black ghettos was far less. “You cannot take any people, of any color, and exempt them from the requirements of civilization — including work, behavioral standards, personal responsibility, and all the other basic things that the clever intelligentsia disdain — without ruinous consequences to them and to society at large. “Non-judgmental subsidies of counterproductive lifestyles are treating people as if they were livestock, to be fed and tended by others in a welfare state — and yet expecting them to develop as human beings have developed when facing the challenges of life themselves.”
Thomas Sowell also takes up these points elsewhere.
In A Conflict of Visions, Sowell explains the basic conflict here in terms of two contrasting viewstwo contrasting views of human nature, the constrained or tragic view and the unconstrained or utopian vision.
The constrained vision of human nature sees human beings as inherently flawed, limited, and selfish. In order to get humankind to cooperate together in societies, it is necessary to have institutions and practices that govern people, establishing norms or rules for cooperation and crucially, incentives for cooperation and, where appropriate, penalties for unacceptably pathological and antisocial behavior.
The unconstrained vision (same source), on the other hand, sees human nature as more malleable and especially improvable. From this perspective, it is possible to improve human beings, to make them less selfish, to create policies that can transcend tradeoffs.
The conflict, then, is not between compassion and heartlessness, it is between unthinking compassion on the one hand, and reason and moral responsibility on the other. I put it in those terms because the constrained vision does more than apply reason, it also uses reason in service to a moral vision of personal responsibility.
While there is a valid place for charity, and different individuals and organizations can decide what that place is and where it begins and ends, it is not “compassionate” to excuse people from the responsibility of taking care of themselves: it is pathological, and the indolence and idleness it incentivizes begets more pathology.
Ask yourself this, is it compassionate or loving to buy alcohol for the alcoholic, heroin for the heroin addict, or donuts for the morbidly obese? Or is it perhaps pathological, a way of enabling harmful and self-destructive behavior?
If you think this analogy to addiction is unfair, consider the practical consequences of relieving adult human beings from the responsibility of being adults: they become removed from the pool of people who are expected to support themselves, and as such become a kind of social and political livestock or even pets. The behavioral and social consequences of this are ruinous.
For retired prison doctor and psychiatrist (and brilliant social critic) Theodore Dalrymple, the pernicious effects of the expansive British welfare state are apparent in the share of the population that has become unfit for any kind of productive activity:
“[…] Britain has enormous cultural problems, perhaps only to be expected in a country in which more than fifty per cent of children are born out of wedlock and twenty per cent do not eat a meal with another member of their household more than once every two weeks. A dangerously high and perhaps unsustainable proportion of the population is unfitted for productive life in a modern economy, having attained an abysmally low educational level despite (or because of?) considerable state expenditure. This section of the population is not merely indifferent to refinement of any kind – intellectual, aesthetic or of manners – but actively hostile to it. Similarly, it is not merely not anxious to learn, it is anxious not to learn.”
Commenting on Dalrymple in The American Conservative, Tom Dreher sees many parallels with the United States:
“A white friend of mine taught in a predominantly black school in a rural Louisiana parish. After several years, she despaired of it, and transferred to a more mixed-race, middle-class school. She told me that nothing she did could break the shackles of poverty culture on the minds of the kids there. These children were not born stupid, but they were deformed by a local culture that disdained education and hard work. Most of the girls — these were ninth graders — aspired to nothing more than having a baby by a boy. Most of the boys aspired to nothing at all. “It’s not just a race thing. I have a white working class friend from that same parish. She’s pretty much the only functional member of her sprawling family. The stories she tells about the laziness, the substance abuse, and the jaw-dropping instability of her clan beggar belief.”
Omni-Compassionism produces votes for politicians, good feelings for the well-intentioned but naïve left-wing voters who vote for more welfare state, and handouts for freeloaders. All of these things incur costs in the form of lost productivity and lost human potential. The welfare state extracts resources from the productive and uses them to subsidize laziness and indolence in a significant share of the population. As we have seen, the personal and social consequences of these policies are ruinous, and they also seem to be dysgenic.
The boundaries of Omni-Compassionism are as revealing as the nature of the beast itself. Leftists are of course highly selective: they’re happy to ‘compassionately’ vote for other people’s money to be spent on the welfare state, but as we’ve seen, they’re often not so compassionate to their political foes.
An excellent example of this comes from a recent episode of The Public Space, in which co-hosts Jean-Francois “JF” Gariépy and Lauren Rose interviewed journalist Faith Goldy on the subject of the recent attack she suffered at the hands of Antifa.
Canada has been suffering a massive influx of migrants since Justin Trudeau’s ill-conceived invitation, many of them coming from or through the U.S. Once in Canada, they are eligible for welfare benefits, including access to Canada’s universal healthcare system.
Goldy, of course, is a nationalist, while the Antifa thugs were demonstrating in favor of open borders—one even told Goldy they did not recognize such a thing as a “Canadian people.” She describes being literally spat upon and physically assaulted by people who no doubt saw themselves as embodying “compassion” for the migrants who have been beating a path to Canada’s doorstep in order to take advantage of the lavish welfare state of that country.
This is the ultimate dramatization of the conflict between the constrained, tragic vision of human nature and the unconstrained, utopian vision, one which reveals the utter hypocrisy of the latter. In the end, those who seek to build an Omni-Compassionate Utopia are not only indifferent to the consequences of their ideology, but also eager to attack anyone who dares to resist them.
(function(w,d,s,i){w.ldAdInit=w.ldAdInit||[];w.ldAdInit.push({slot:10817587730962790,size:[0, 0],id:"ld-5979-7226"});if(!d.getElementById(i)){var j=d.createElement(s),p=d.getElementsByTagName(s)[0];j.async=true;j.src="//cdn2.lockerdomecdn.com/_js/ajs.js";j.id=i;p.parentNode.insertBefore(j,p);}})(window,document,"script","ld-ajs");
The dirty secret of the left is that Omni-Compassionism, all the Universalism of the progressive ideology, turns to anger, moral outrage, and character assassination the moment someone dares to dissent. From the de-platforming of speakers to the demonization of anyone wearing a MAGA hat to calls to deport White people, the Religion of Compassion is outright hateful to those it deems to be heretics, apostates, and unbelievers.
This, then, is liberal compassion unmasked: selectively indulgent compassion, in blind ignorance of the pathological consequences, and demonization of anyone who dares to point them out.
Thank you for reading Republic Standard. We publish this magazine and the Freebird Forum because we believe in free speech- but it doesn't come cheap! Will you make a small donation towards our running costs? You can make a difference by clicking here.
If you love free speech, we are building the platform for you! Read about how we are building FreebirdTV, open source video hosting with no thought-policing.
The Republic Standard Web Shop is now open! Every piece of merchandise you buy is a victory against the nerds.
from Republic Standard | Conservative Thought & Culture Magazine https://ift.tt/2GJi5vv via IFTTT
0 notes
Text
Thoughts on Healthcare Reform
Original Post from 2009, with updates.
The single most problematic issue with our healthcare system is the purchasing structure. There is no way we would have ended up with a system that features one party purchasing the equivalent of a service plan and extended 3P warranty (note: not an insurance plan) on behalf of the one who would ultimately consume the services if it weren't for some historical distortions involving price controls, and then incentives written in to the current tax structures. The consumer is two steps removed from caring about the ultimate price of the services, or having any power to really negotiate anything. [UPDATE: See http://www.businessinsider.com/heres-why-health-insurance-is-so-weird-2013-11 on the strangeness of health "insurance."]
Again, it's not insurance, generally. And to the extent we move to disallow exclusions for pre-existing conditions, it ceases to be insurance at all. But this may not be a bad thing, given the progress of science and genetics in identifying what disease will likely kill you (minus a car wreck or an encounter with a great white shark or something). See:http://blogs.ft.com/maverecon/2009/07/the-inevitable-socialisation-of-health-care-financing/
Note that this has nothing to do with a profit motive in insurers. It's the basic economics of insurance. [UPDATE: Note this liberal economic critique of Obamacare: https://baselinescenario.com/2016/05/09/the-problem-with-obamacare/ ]
We really need to separate out medical care into 5 baskets (maybe the first two could be combined; same with the last two): 1) Real public goods, like vaccinations; 2) basic diagnostic and preventative care (if the government is picking up the tab, this essentially included in 1); 3) Basic medical care; 4) Catastrophic injury care and rehab; or 5) long-term deadly disease care.
In my mind, the government should pay for 1 for sure; I could be convinced that it should also pay for 2; people should pay for 3 themselves; and insurance should be purchased or a socialized risk pool created to cover 4 and 5. To the extent genetics advances to the point of eliminating uncertainty on diseases, this may need to be socialized, which would essentially mean that instead of insurance people would pay into a government-controlled pool that is set up like something akin to insurance.
We should look closely at two systems that for some reason people don't seem to be talking about nearly as much as Canada's and Europe's:
Australia: http://brontecapital.blogspot.com/2009/08/health-care-reform-and-single-payer.html
and Singapore: http://www.themoneyillusion.com/?p=2974 ; http://www.american.com/archive/2008/may-june-magazine-contents/the-singapore-model
I'd lean toward Singapore.But a hybrid system with Australian-style hospitals and instacare facilities for basic and also to provide basics for the indigent undergirding a Singapore style could work really well.
[UPDATE: Some people have told me that they don't think we could do an Australian-style system because of how much it would cost to build the hospitals. I would agree - if we had to build them all from the ground up. There are already a whole lot of government-owned and run hospitals in the U.S., and it may well be more efficient and better to put them all under central management in service of this type of system. The first group is the VA hospitals. The second group are those government hospitals that are run by local governments (city and county hospitals) and state governments (particularly state-run university hospitals). Starting with that as a base, we could definitely get something going pretty quickly.]
[UPDATE: A good post on why it's culturally difficult for Americans to agree on healthcare reform: http://www.arnoldkling.com/blog/the-cultural-roots-of-americas-health-care-policy-mess/ ]
[UPDATE: Tom Hardman put together a great summary of the main ways other countries have enacted universal (or near-universal) coverage here: https://tomhardmanblog.wordpress.com/2017/03/11/health-care-seeing-both-sides/
Quoting Tom:
"All other wealthy, technologically advanced, industrial democracies in the world guarantee a basic level of medical care to anyone who gets sick, while spending far less than the U.S. does on health care.
There are three basic models that countries use to provide universal coverage:
—The Bismarck Model: This model uses private health care providers (doctors, nurses, hospitals, etc.) and private health insurance companies. All citizens are required to purchase health insurance, the government exercises tight regulatory control over insurance coverage and pricing, and health insurance companies are non-profit entities. This kind of model is found in Germany, France, Japan, Switzerland, the Netherlands, and Belgium. (Obamacare is quite similar to the Bismarck model, but Obamacare’s individual mandate is much weaker than in other countries that use this model.)
—The Beveridge Model: This is probably what Americans have in mind when they talk about “socialized medicine.” With this model, health care is provided and financed by the government, through tax payments. There are no medical bills, insurance premiums, or co-payments; rather, medical treatment is a public service, like the fire department. Most hospitals and clinics are owned by the government, and most doctors are government employees. Countries using the Beveridge Model, or variations on it, include Great Britain, Italy, Spain, Scandinavian countries, and Hong Kong.
—The National Health Insurance (Single-Payer) Model: This is the kind of health insurance system that Bernie Sanders was proposing: the providers of health care are private, but the payer is a government-run insurance program that every citizen pays into. The paradigmatic single-payer system is Canada’s; Australia, Taiwan, and South Korea have adopted variations on the single-payer model."
Singapore is a hybrid model based on The Bismark Model, but with improvements. It's still my favorite - see: https://www.nytimes.com/2017/03/18/opinion/sunday/make-america-singapore.html ]
And I still would have worries about impacting medical innovation ( http://www.futurepundit.com/archives/006455.html ; http://www.washingtonexaminer.com/opinion/columns/OpEd-Contributor/The-hidden-cost-of-national-health-care-7952906-50469092.html ), but I find it increasingly difficult to justify forcing American consumers to subsidize foreign consumers of medical technology, including both drugs and devices. [UPDATE: See http://time.com/money/4130688/prescription-drug-prices-america/ and also http://www.ibtimes.com/how-us-subsidizes-cheap-drugs-europe-2112662 ][UPDATE: Some more claims on high drug prices being the result of some regulations http://www.economist.com/blogs/economist-explains/2016/09/economist-explains-2?fsrc=scn/fb/te/bl/ed/whydrugpricesinamericaaresohigh ][UPDATE: Another factor in high pharmaceutical costs: oligopoly among prescription benefit manageement companies: http://prospect.org/article/hidden-monopolies-raise-drug-prices-0 ]
Some more ideas: I'd increase the supply of doctors - the AMA's cartel both keeps out highly qualified foreign practioners and also won't increase the number of accredited schools producing doctors here at home. Additionally, there should be transparent and up-front pricing for all services, to allow for price discrimination and competition on price in addition to quality. [UPDATE: See here for a few more supply-side ideas: http://marginalrevolution.com/marginalrevolution/2017/03/supply-side-health-care-reform.html ]
I'd increase the non-hospital urgent care facilities for check ups for kids with fevers and broken arms and such - kind of similar to Australia's system of public hospitals. It might make sense just to build those. You could start with using the VA and academic hospitals - maybe state and local government-run hospitals, too.
If we need to keep insurance, then I'd nationalize (as opposed to state-level that is) insurance regulation, including creation of standards for minimum insurance and create a true catastrophic-only policy.
Oh, and lest I forget: reform malpractice law.http://correspondents.theatlantic.com/philip_howard/2009/08/stonewalling_legal_reform.php
[UPDATE: And don't let anyone try to claim that there isn't a sizable amount to be saved just from reforming billing practices and forcing providers to give upfront pricing. We spend 25% of our healthcare total on administrative costs, many relating to coding for billing. See https://www.nytimes.com/2017/03/29/magazine/those-indecipherable-medical-bills-theyre-one-reason-health-care-costs-so-much.html ]
We should look to improve; we should be wary about messing up the good points of what we have. [UPDATE: And, contra many stories, what we have is pretty darn good - for example, life-expectancy, after controlling for accidents and violence (which are not really health-care related), puts the U.S. at the top: https://www.adamsmith.org/blog/us-healthcare-most-people-dont-know-what-theyre-talking-about ]UPDATE: And, we already spend a more on healthcare for the poor than on healthcare for the rich: http://marginalrevolution.com/marginalrevolution/2017/06/americans-spend-money-health-care-poor.html ]
One final point: There is no paradox if you hold the two following beliefs simultaneously: 1) You think there are issues with the current system that could be improved; and 2) You don't want to support something you think will make things worse overall.
Here are some other ideas of what we could do instead of Obamacare:
[UPDATED: http://marginalrevolution.com/marginalrevolution/2013/06/options-for-health-care-reform.html
https://www.nytimes.com/2017/03/21/opinion/how-trump-can-fix-health-care.html?_r=0]
http://www.marginalrevolution.com/marginalrevolution/2009/11/what-should-we-do-instead-of-the-health-reform-bill.html
Here are some other good posts or articles thinking about the issues:
http://www.marginalrevolution.com/marginalrevolution/2007/10/how-to-debate-h.html
http://www.willwilkinson.net/flybottle/2006/03/19/health-care-fantasia/
http://meganmcardle.theatlantic.com/archives/2009/09/practical_philosophy_again.php
http://www.theatlantic.com/doc/200909/health-care
http://keithhennessey.com/2009/08/28/who-should-decide-part-1/
http://www.hoover.org/publications/digest/49525427.html
http://meganmcardle.theatlantic.com/archives/2009/07/a_public_plan_and_the_law_of_u.php
http://meganmcardle.theatlantic.com/archives/2009/07/extreme_health_care.php
and
http://daylightsmark.blogspot.com/2009/07/health-care-those-who-are-not-liberal.html
Those are my thoughts - anyone have any better ideas?
0 notes
Text
Part 6: Paying for Denmark's Free Health Care
TL; DR Summary
Part 1: Denmark is the world’s happiest nation because of the following reasons (Not true)
Part 2: Denmark has a $20/hour minimum wage (No, it does not).
Part 3: Denmark has a 33 hour work week (No, it does not)
Part 4: Denmark has Free University ? Yes but read why – it’s not why you think)
Part 5: Denmark has Free Childcare? No, it does not – this claim is a lie).
Part 6: Denmark has Free Healthcare, sort of
Part 7: We should all be like Denmark, remember?
Part 8: Is Denmark a socialist country?
This final installment looks at the claim that Denmark has free health care – health care is free at point of service, but obviously, health care expenses still get paid some how.
FREE HEALTHCARE
Clinics are free. However, (as of 2012) there was an 8% flat rate “Health contribution” tax on income. This tax was being phased out and rolled in to general income taxes. However, this gives us a baseline for what “free” health care costs in Denmark. Additional tax revenue, mostly from local government income taxes, is added to this to pay the costs of “Free” health care.
The 8% health care income tax applied to all earnings above the US$7600 level. Starting in 2012, this is going down by 1% per year and merged into the general income tax (Source Danish government web site reviewed in 2015. For additional reference, see also http://www.civitas.org.uk/pdf/Denmark.pdf)
Update 2019: See also “Healthcare in Sweden” and “Swedish Healthcare“)
TAXES IN DENMARK
Taxes in Denmark are the highest in the world. The typical Dane worker pays between 50% and 70% of their income, as taxes, including “Gross”, “Health”, “Income”, and “Local” taxes, plus a 25% Value-Added-Tax (VAT) which is similar to the U.S. concept of a sales tax. The “50-70%” figure is NOT a marginal tax rate – that is the total percentage of one’s income that is paid in taxes. (In Scandinavian countries, most of the tax burden falls on the broad middle and lower class as flat rate social security, Value Added Tax and payroll taxes, and high municipal or local government taxes.)
The basic tax scheme in Denmark is roughly:
8% Gross tax (applies to all income, starting with the first earnings)
8% Health contribution tax (being merged in to the income tax)
6-15% National income tax
24-36% local income tax
25% Value-Added Tax on most purchases made with your after tax income
Additional taxes such as a 150% tax on the purchase of an automobile (just reduced from 180% – plus other annual taxes like fuel taxes).
Sources for Denmark tax information:
http://taxindenmark.com/article.31.html
http://www.civitas.org.uk/pdf/Denmark.pdf
http://www.fool.com/investing/general/2013/04/14/think-your-taxes-are-high-the-5-countries-with-the.aspx
The 8% Health contribution tax that existed to 2012 did not fully fund the Danish health system. Local government provides additional funding from local income taxes.
While health clinics provide free medical care, dental care and vision care are not free. Prescriptions are not entirely free either. These additional expenses are paid out of pocket or via insurance benefit programs.
DOCTOR TRAINING AND PAY
“Doctors” in the Danish system earn a Bachelor + “Candidate of Medicine” degree (equivalent to a U.S. Masters degree) in a combined total of 6 years. In the U.S., “Doctors” earn a 4 year undergraduate degree (Bachelor) and then a 4 year Medical Doctor degree (combined total of 8 years). Those with a “Candidate of Medicine” degree may optionally choose to pursue a dr.med. or PhD degree; both are typically research degrees. Many doctors may choose to pursue additional training and specialty certifications. These differences should be considered when comparing the two health care systems. Many U.S. Medical Doctors also obtain additional years of training beyond the 4 years of medical school.
Danish doctors are paid less than U.S. doctors, on average. According to a relative who is a U.S. medical doctor, most Danish doctors have a standard work week, unlike U.S. doctors who often work 55-60+ hours per week, plus being on call at various times. A consequence is that US doctors are paid more, on average, but also work more hours.
MDs per 1,000 patients
In discussions comparing U.S. health care to that in other countries, a common metric introduced for comparisons is the number of doctors per 1,000 patients. A common claim seen on social media is that “Germany, France, UK, Canada, Australia, Japan have more MD’s per 1,000 patients”.
First, we look at data from the World Health Organization, and then we look at the definition of “doctor” – as there are differences in the training of “doctors” by country.
Country, MDs/1000 and Mortality rate, sorted by the MDs/1000 value:
Germany 3.889, 11.29
Australia 3.273, 7.07
France 3.19, 9.06
U.S. 2.451, 8.15
Japan 2.297, 9.38
Canada 2.068, 8.31
Sorted by mortality rate:
Australia 3.273, 7.07
Canada 2.068, 8.31
U.S. 2.451, 8.15
France 3.19, 9.06
Japan 2.297, 9.38
Germany 3.889, 11.29
Data comes from the World Health Organization. Mortality rates also depend on other demographic factors including age distribution and common health issues that may be unique to each country.
The Definition of Doctor Depends on the Country
Doctors in the U.S. earn a medical doctor degree as their terminal degree, while in other countries, a “doctor” may complete a training program (sometimes equivalent to the U.S. and sometimes not).
A U.S. doctor has 8 years of training, a German doctor has 6 years of training, and in the U.K., a medic initially has 5 years of training. In many countries, physician candidates go on to more training, such that their training is similar to that in North America.
In Germany, a medical degree is a total of 6 years undergraduate training (2 years basic sciences, 3 years clinical, 1 year practical). All tuition is free in Germany – versus around $300,000 in the U.S. just for the 4 years of medical school.
In the UK, a medical degree is a total of 5 years of training (2 years foundation courses, 3 years general practice) resulting in a “Bachelor of Medicine” or “Bachelor of Surgery” degree and becomes a general practitioner. Additional training may be undertaken, adding more years to the training. Until 1998, tuition was free. Now students pay tuition similar to our public universities.
In France, a medical degree is 6 years. Tuition is subsidized by the government.
In Australia, students enter directly out of high school for 5 or 6 year training programs (both the 5- and 6-year programs exist there). Tuition cost is a fraction of U.S. tuition cost (perhaps 1/4 to 1/3d).
In Japan, students enter directly out of high school equivalent and enter a 6-year program. Tuition is similar to a US public university.
Canada is the same as the U.S.
Comparing the number of MDs per 1,000 across countries might not be a good metric for comparing the health systems between countries since not all “doctors” are equivalent.
U.S. Politics and the Danish Health Care System
The creation of a single-payer health care model in the U.S. was widely promoted by candidate Bernie Sanders during the 2016 Presidential campaign.
Much of what he offered, however, was lacking in details. Reporter Ezra Klein wrote that Bernie Sanders’ single payer health care proposal did not make sense in a detailed review of what Sanders proposes.
Other posts in this series
Part 1: Denmark is the world’s happiest nation because of the following reasons (Not true)
Part 2: Denmark has a $20/hour minimum wage (No, it does not).
Part 3: Denmark has a 33 hour work week (No, it does not)
Part 4: Denmark has Free University ? Yes but read why – it’s not why you think)
Part 5: Denmark has Free Childcare? No, it does not – this claim is a lie).
Part 6: Denmark has Free Healthcare, sort of
Part 7: We should all be like Denmark, remember?
Part 8: Is Denmark a socialist country?
Note – All data source links used were re-checked in January 2017 and were “live”.
Notes on Single Payer
Note – I am not opposed to a single payer scheme. I need to add this because of the logical fallacy that if I criticize or question X, then I obviously approve of Y, when I have not said any such thing.
To elaborate, a problem with ObamaCare is it failed to control costs in the non-group market, which was one of the primary goals for which ObamaCare was created. Rates in the non group market skyrocketed in significant part because the insurance “risk pools” are too small, coupled with the non-group risk pools absorbing all of the risks (costs) of the previous state-run high risk insurance pools, and other new risks. As of 2017, these risks are not shared with any other insured parties – consequently, rates in the non group market shot up so high that many can no longer afford to buy ObamaCare health insurance.
Our own ObamaCare health insurance premiums rose by 140% from 2014 to 2015 to 2016 to 2017.
Using HealthCare.gov, I collected price quotes around the U.S.
For many scenarios, the price of a basic “Silver” plan runs families $2,000 to $3,000 per month. For 2017, an age 64 couple living in Flagstaff, AZ, and earning $64,000/year in income – just above the subsidy level – has a nearly $3,000/year or $36,000/year health insurance bill. The same 64 year old couple has a bill of over $2,500/month in Asheboro, North Carolina or $4,000/month in Homer, Alaska. In Klamath Falls, Oregon, a 52 year old couple with 3 kids has a monthly bill of about $2,000/month. No one can afford to pay these rates.
Actual screen shot from HealthCare.gov for Flagstaff Arizona for 2017:
ObamaCare insurance is not like corporate insurance plans but leads the way in high deductibles (up to nearly $7,000 per person per year depending on plan and market) and narrow provider networks (we had access to some providers in 3 counties only – all other access while traveling, for example, has up to a $14,000 per person per year deductible and is billed at higher “out of network” rates.) Literally, in a worst case situation, a family could have pay $24,000 per year for health insurance, but then not collect anything until paying an additional $7,000/person out of pocket for the deductible. While not likely, the 52 year old couple with 3 kids could have to pay $55,000 out of pocket in one year before collecting benefits.
What happened? Due to defects in the Affordable Care Act, high risks were pushed into the small individual insurance risk pools. The individual (or non group market) was turned into a high risk, high cost insurance pool, driving rates sky high for much of the non group market.
What happened to the patients in the 35 state run high risk pools? Almost all were rolled into the non-group market place since “pre-existing conditions” were no longer excluded and prices are based solely on age, location and whether you smoke. While these patients made up 2% of the non-group market, they are extraordinarily expensive. As of 2017, all of their risk (cost) is borne solely by the other members of the small individual market risk pools. There are additional examples of other risks that were pushed into the small non group market, resulting in unaffordable insurance rates.
Consequently, the non-group market was turned into a high risk pool with rates. As of the end of 2016, the size of the ACA nongroup marketplace was less than half what was expected and too small to distribute the risks (costs).
There are several ways to address this failure which concentrated all of the high cost patients into the non-group market. One way is a single payer model as that would spread the risk (cost) to the entire population.
If you wish to learn more, I have written an entire paper, with numerous cited references, that explains the problem and potential solutions in far greater detail. Unfortunately, 100% of my elected representatives ignored this input. Due to their being stuck on ideological stupidity, no elected official in the US seriously wants to fix these problems. They – regardless of which side – were interested solely in ideology and not addressing the problems faced by real people – their own constituents. Literally, they did not care.
More
Part 1: Denmark is the world’s happiest nation because of the following reasons (Not true)
Part 2: Denmark has a $20/hour minimum wage (No, it does not).
Part 3: Denmark has a 33 hour work week (No, it does not)
Part 4: Denmark has Free University ? Yes but read why – it’s not why you think)
Part 5: Denmark has Free Childcare? No, it does not – this claim is a lie).
Part 6: Denmark has Free Healthcare, sort of
Part 7: We should all be like Denmark, remember?
Text for Search Indexing
Why is the Denmark ranked the happiest country in the world by the United Nations? $20 minimum wage 33 hour work week Free University Free Child Care Free Health Care
Why is Denmark the happiest country in the world? $20 minimum wage
33-hour work week
Free university
Free childcare
Free healthcare
Share if America should follow their lead
Occupy Democrats
Denmark v. USA $21/hr. minimum wage $7.25/hr. minimum wage Free healthcare, childcare, college and job training – Healthcare, childcare and college are a luxury, can bankrupt you or saddle you with debt Paid sick and parental leave – No paid sick/parental leave Only 6.1% of children live in poverty – 23.1% of children are poor, highest rate in rich world Ranked #1 happiest country Ranked #1 country for business Ranked #1 most unequal rich country Share if Americans can learn from Denmark! Occupy Democrats
Part 6: Paying for Denmark’s Free Health Care was originally published on SocialPanic
0 notes
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://venabeahan.tumblr.com
0 notes
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://wittooth.tumblr.com/
0 notes
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://wittooth.tumblr.com/
0 notes
Text
The Medical AI Floodgates Open, at a Cost of $1000 per Patient
By LUKE OAKDEN-RAYNER
In surprising news this week, CMS (the Centres for Medicare & Medicaid Services) in the USA approved the first reimbursement for AI augmented medical care. Viz.ai have a deep learning model which identifies signs of stroke on brain CT and automatically contacts the neurointerventionalist, bypassing the first read normally performed by a general radiologist.
From their press material:
Viz.ai demonstrated to CMS a significant reduction in time to treatment and improved clinical outcomes in patients suffering a stroke. Viz LVO has been granted a New Technology Add on Payment of up to $1,040 per use in patients with suspected strokes.
https://www.prnewswire.com/news-releases/vizai-granted-medicare-new-technology-add-on-payment-301123603.html
This is enormous news, and marks the start of a totally new era in medical AI.
Especially that pricetag!
Doing it tough
It is widely known in the medical AI community that it has been a troubled marketplace for AI developers. The majority of companies have developed putatively useful AI models, but have been unable to sell them to anyone. This has lead to many predictions that we are going to see a crash amongst medical AI startups, as capital runs out and revenue can’t take over. There have even been suggestions that a medical “AI winter” might be coming.
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating… pic.twitter.com/FMHfdt6lNT
— Curt Langlotz (@curtlanglotz) November 1, 2019
To be clear, this was never a problem with the technology. Deep learning works, and there are lots of ways it can be applied usefully in medicine. It was an alignment problem: the people who procure medical technology (typically CIOs) are motivated by business needs, not how useful a model is.
The strongest business incentive is money, earning more or spending less, and proving that AI models can help here has been really difficult.
Most researchers and developers have focused on medical outcomes, like diagnostic performance or lives saved. But even if a model saves lives, it might not impress a CIO because healthcare providers have no inbuilt incentive to help people. Gross, I know, but medicine is full of perverse incentives. Nations and employers care about health and wellbeing (because it improves productivity and is generally popular among constituents), but hospitals (both public and private) care about something else.
They care about reimbursement.
Money, that’s what I want
Reimbursement is how medicine incentivises actually helping people. A central payer, whether a government or an insurance company, decides what medical management is cost-effective to improve health.
When a test or treatment is reimbursed, then healthcare providers get paid to use it. All of a sudden, CIOs are really excited. Pay some money to a company, get as much or more money back for using the product.
Does it work?
Well, I’ve spoken about mammography CAD before, an old form of AI intended to assist in detecting breast cancer. This became popular in the 00s, when CMS decided to reimburse CAD-aided mammography tests. A provider would get about $10 more if they used CAD than if they did “standard” reading.
Within a decade almost every screening mammogram in America is read with CAD assistance.
No, it isn’t viral infection rates, yes it was a decade of exponential growth. Unrelated fact: mammography CAD was first reimbursed in 2001.
But, you say, maybe they just used it because it was amazing?
Nope. It didn’t work.
How well CAD worked in practice (~500,000 patients), from Lehman et al.
In fact, nobody else uses it. I’ve never found the exact numbers, but CAD use outside the USA is practically non-existent. Why? Cos it doesn’t work, and you don’t get paid for it.
Just think about that. Medicare has spent hundreds of millions (billions?) on a technology which didn’t work, driving widespread use. Financial incentives are powerful and dangerous things*.
Time is brain
Things happen in the brain. Over time. Electric.
So, financial incentives are the big deal. Life or death for new technologies. So far, modern medical AI (by which I mostly mean deep learning) has received dozens of FDA clearances, but there has been almost no financial incentive to use these products.
So what is ContaCT, and how did Viz.ai get CMS to reimburse its use?
Viz.ai received FDA clearance in early 2018 for a deep learning system that can detect blockages in the large blood vessels that supply the brain, on CT scans. This system was an interesting break from the dozens of pure diagnostic systems that startups were producing at the time, in that it was intended purely for triage and fast response. If it saw a blockage, it directly contacted the specialist who could fix the problem, skipping the radiologist who would normally read the image first.
Viz.ai claim that by reducing the time for a specialist to review the CT scan of possible blockages, they prevent long delays during which time more and more brain cells are dying from a lack of blood. They have published a few papers on the topic (here and here) and had to provide a fair bit more to CMS to justify this claim.
The CMS document that describes the decision to reimburse ContaCT is 40 pages long, but is well worth a read if this whole topic is of interest. There is a lot in there, with a lot of back and forth between CMS and Viz.ai, covering a lot of topics (including many that I have seen raised on Twitter). I’ve uploaded the document here (extracted from a longer 2000+ page document on other CMS decisions).
CMS requires that applicants prove the technology produces “substantial clinical improvement”. So what did Viz.ai provide?
They show several things:
faster time to notification of the clot-busting specialist
faster time to transfer from peripheral hospital to a central hospital where the relevant procedure can be performed
faster time to clot-busting procedure
These things alone are interesting, but rely purely on our existing knowledge that delays lead to worse brain injuries (as the saying goes, “in strokes, time is brain”). But Viz.ai didn’t stop there. They actually did the thing I always harp on about. They showed outcomes.
Improved modified Rankin score (mRS) at discharge
Improved NIH Stroke Score (NIHSS) at day 5
Improved mRS at day 90
These outcomes show that these patients did better than patients without ContaCT. These scores are widely used in stroke trials and summarise degree of damage/disability following a stroke**. So that is awesome, finally we have evidence of a clinical improvement for a radiology AI system!
Limitations
Not everyone was impressed with the evidence provided to CMS when this story hit the webs.
I don’t care if the results were “significant” but n=43 patients doesn’t say much, even if prospective.
— Ahmed Hosny (@ahmedhosny) September 4, 2020
Hugh and Ahmed raise two main points.
that the time-saving comes from cutting out the radiologist and getting the neuro-interventionalist to review the CT scan directly
that the sample size is pretty small
I’m going to take off my skeptical hat and disagree with both of them!
Several people argued that this isn’t actually an AI intervention at all (or even a technological one), and that all they are doing is changing the care pathway. I find this claim dubious – it relies on the idea that stroke management would be better if a neuro-interventionalist (abbreviated to INR from here on) read every CT angiogram performed for a possible stroke.
There is a problem with this. INRs are rare. These are subspecialists. In my state in Australia, population 1.7 million, we have four of them. Across the whole of the US, there were previously estimated to be 200-400 INRs, although those figures are quite old.
In the US there are over 1,000,000 stroke admissions per year (~850,000 from Medicare alone in 2010). There is no way these busy INRs can review all those scans.
This is where the AI comes in. If the AI picks up a possible blockage, the INR is contacted. According to Viz.ai, their ContaCT system detects ~90% of blockages, and will exclude around 90% of the patients who don’t have a blockage. So instead of reviewing a million scans a year, the INRs only need to review 100,000. Much more achievable with the limited workforce.
So, yes, the innovation here is that the INR sees the scan before the radiologist, but it only works because the AI system cuts out the majority of the scans.
Then we come to the complaint about sample size. I’m normally all about criticising studies for small sample sizes, and it is true their clinical outcomes results (the mRS and NIHSS results) were in 43 patients. But they did provide a lot more data on the other outcomes. Across 3 additional sites, they show that another 80 or so patients had a statistically shorter time to puncture than controls. They also show that their entire database of real-world cases where ContaCT was used, almost 5,000 patients, achieved the same time-to-notification as they had in the original study.
In combination, these results are all reassuring. It is also worth noting that they are currently carrying out a large multi-centre study and we will see a larger sample size for the outcomes results in the near future. Sure, I would prefer to see that before reimbursement, but I’m not shocked that the decision was made this way.
Seriously though. $1000?
The announcement that Medicare would reimburse providers up to $1000 per use of the AI model was by far the most controversial part, and for good reason. AI models cost pretty much nothing to run. The CT scan itself, which determines if a patient can be treated, is about $1000. Does CMS think that this AI is as useful as CT scanning in stroke?
Well, no. Of course not.
This whole thing is a bit weird, but essentially CMS have tried to work with the business model of Viz.ai, which is unlike any other medical technology business model. Viz.ai charge a yearly subscription to deploy and maintain their AI system.
I don’t know the actual pricing for ContaCT, but the document repeatedly refers to a cost of $25,000 per annum. In this example, they say that the reimbursement cost is designed to cover the subscription. If there are 25 patients in the year, then they reimburse $1000 per patient. If there are 500 patients (much more likely), they reimburse $50 per patient.
Do note that the actual payment seems to be fixed for a year at a time. So it is absolutely true that in 2021 each user of ContaCT will be reimbursed $1000. At the end of the 2021 financial year CMS will look at how many claims there were and revise the payment (down, presumably). So it is possible that some high volume hospital will make out like bandits this year and be reimbursed a million dollars for a 25k subscription (ie if they use the AI system 1000 times). Not sure if there are safeguards against that.
The reason the press release is talking about $1000 dollars is that this is the cap on reimbursement per patient. So if a hospital scans less than 25 patients per year, they cannot recoup all their costs and will be out of pocket.
If anything, this approach is conservative. No matter how much the system is used, no matter how much value it generates, it only costs $25,000 per year. This is not the runaway profit that many imagined for medical AI (although broad coverage of hospitals would still be incredibly lucrative).
What this model does do, however is produce guaranteed revenue, which is a huge step forward in this challenging space. Winter is averted, maybe.
Closing thoughts
This is a massive deal.
I hadn’t mentioned this yet, but honestly I didn’t see this coming and neither did many others I have spoken to. I thought we were probably years away from reimbursement for AI, and that it would probably start in mammography.
Wow! I must admit I did not think this would ever happen: CMS approves payments for https://t.co/aJIvpjHWjD software https://t.co/7pyrOksAwC
— Curt Langlotz (@curtlanglotz) September 4, 2020
While the exact funding mechanism is a bit strange, startups now have a clear path to follow to generate revenue. If this doesn’t stabilise the market, I don’t know what could.
That isn’t to say it will be easy to follow Viz.ai’s footsteps here. It still remains to be seen if this decision by CMS will translate into widespread adoption (and this may hinge to some extent on the results of their large trial). It is also true that Viz hit on a formula here which is unusual. This new pathway works because there is no obvious risk – if the model misses a stroke they still receive the current standard of care, which is review by a radiologist. Thus far I haven’t been able to come up with another use case that this would work in. Maybe if you can think of one, let me know in the comments or on Twitter.
But even if it is not a simple path to follow, at least it is a path. I am certainly re-evaluating my expectations as well. More reimbursements, for less restrictive tasks, may be just around the corner.
This is where it started, folks.
* For an interesting little discussion on how this happened, Joshua Fenton summarised the extensive lobbying effort led by silicon valley Congresswoman Anna Eshoo in an editorial (unfortunately paywalled) for JAMA, where he asked if we should stop spending 1 in every 10,000 dollars in US healthcare on a failed technology. Spoiler: we still do.
** The mRS and NIHSS scores aren’t perfect by any means, but are pretty broadly accepted as endpoints for this sort of study.
Luke Oakden-Rayner is a radiologist in South Australia, undertaking a Ph.D in Medicine with the School of Public Health at the University of Adelaide. This post originally appeared on his blog here.
The Medical AI Floodgates Open, at a Cost of $1000 per Patient published first on https://wittooth.tumblr.com/
0 notes
Text
As I’ve always suspected, Health Care = Communism + Frappuccinos
By MATTHEW HOLT
Happy 15th birthday THCB! Yes, 15 years ago today this little blog opened for business and changed my life (and at least impacted a few others). Later this week we are going to celebrate and tell you a bit more about what the next 15 years (really?) of THCB might look like. But for now, I’m rerunning a few of my favorite pieces from the mid-2000s, the golden age of blogging. Today I present “Health Care = Communism + Frappuccinos”, one of my favorites about the relationship between government and private sector originally published here on Jan7, 2005. And like the Medicare one from last week, it sure holds true today. Matthew Holt
Those of you who think I’m an unreconstructed commie will correctly suspect that I’ve always discussed Marxism in my health care talks. You’d be amazed at how many audiences of hospital administrators in the mid-west know nothing about the integral essentials of Marx’s theory of history. And I really enjoy bring the light to them, especially when I manage to reference Mongolia 1919, managed care and Communism in the same bullet point.
While I’ve always been very proud of that one (err.. maybe you have to be there, but you could always hire me to come tell it!), even if I am jesting, there’s a really loose use of the concept of Marxism in this 2005 piece (reprinted in 2009) called A Prescription for Marxism in Foreign Policy from (apparently) libertarian-leaning Harvard professor Kenneth Rogoff. He opens with this little nugget:
“Karl Marx may have suffered a second death at the end of the last century, but look for a spirited comeback in this one. The next great battle between socialism and capitalism will be waged over human health and life expectancy. As rich countries grow richer, and as healthcare technology continues to improve, people will spend ever growing shares of their income on living longer and healthier lives.”
Actually he’s right that there will be a backlash against the (allegedly) market-based capitalism — which has actually been closer to all-out mercantilist booty capitalism — that we’re seen over the last couple of decades. History tends to be reactive and societies go through long periods of reaction to what’s been seen before. In fact the 1980-20?? (10-15?) period of “conservatism” is a reaction to the 1930-1980 period of social corporatism seen in most of the western world. And any period in which the inequality of wealth and income in one society continues to grow at the current rate will eventually invite a reaction–you can ask Louis XVI of France about that.
But when Rogoff is talking about Marxism in health care what he really means is that, because health care by definition will consume more and more of our societal resources, the arguments about the creation and distribution of health care products and services will look more like the arguments seen in the debates about how the government used to allocate resources for “guns versus butter” in the 1950s. These days we are supposed to believe that government blindly accepts letting “the market” rule, even if for vast sways of the economy the government clearly rules the market, which in turn means that those corporations with political influence set the rules and the budgets (quick now, it begins with an H…). That’s how defense has always been and how pharmaceuticals will increasingly be. Rogoff recognizes the centrality of this argument in his description of what’s wrong with American health care:
“Part of the rise in U.S. healthcare costs stems from the breakdown of the checks and balances that more centralized systems provide. (For example, Americans are several times more likely to receive heart bypass surgery than Canadians, where the procedure is reserved for extreme cases. Yet several studies suggest that patients are no worse off in Canada than in the United States). And even the most fanatical free marketers recognize that healthcare is different from other markets, and that the standard supply-and-demand principles don’t necessarily apply. Consumers have poor information, and there is an obvious case for greater government involvement than in other markets.”
But he then goes on to say that the much greater spending seen here as compared to Canada and the UK creates both a terrible service level (and by implication quality level) and diminishes innovation in health care services. And if all countries squeezed profits in the health sector the way Europe and Canada do, there would be much less global innovation in medical technology.
“Today, the whole world benefits freely from advances in health technology that are driven largely by the allure of the profitable U.S. market. If the United States joins other nations in having more socialized medicine, the current pace of technology improvements might well grind to a halt. Even as the status quo persists, I wonder how content Europeans and Canadians will remain as their healthcare needs become more expensive and diverse. There are already signs of growing dissatisfaction with the quality of all but the most basic services. In Canada, the horrific delays for elective surgery remind one of waiting for a car in the old Soviet bloc. And despite British Chancellor Gordon Brown’s determined efforts to rebuild the country’s scandalously dilapidated public hospital system, anyone who can afford to go elsewhere usually does.”
His conclusion is that because for the sake of social equity government intervention in the system is warranted, the health sector will be a “battleground” between capitalism and socialism through this century. If you get past his mis-use or mis-understanding of the terms “capitalism” and “socialism”, the point he’s making is quite interesting. It does though suffer from a typically Amero-centric bias. Rogoff assumes that the extra spending on health care in America leads to better services and by implication better quality. But that’s an old chestnut. By that measure the higher spending in Canada (11% of GDP) should lead to a better system than in France (9%) or Germany (10%). But in those two nations access to drugs and technology is much greater than in the UK or Canada, and things like waiting times are comparable to the US — in fact in Australia and New Zealand they’re better than they are here. A few years back The Economist said that the Swiss system (again several percentage points cheaper than here) was better than the American on an absolute level. Furthermore recent studies of international care quality suggest that particularly for primary care, the US is results-wise(at best) in the middle of the pack. All of those nations have a heavier proportion of government funding of health care spending than in the US, and all of them spend a whole lot less money. Note that the US government spends more per head (and damn nearly as much as share of GDP) on health care as the whole of the UK.
So that all tells us one thing. We’re paying a lot more for health care here, but it isn’t necessarily getting us better outcomes, innovation or even services. We might though have nicer waiting rooms and we certainly lead the league in surgeons with Porsche 911s. Therefore it’s a stretch to imply that higher private spending leads directly to innovation and better services, particularly if the system is not set up with either government-based or real market-based co-ercive capabilities to promote efficiency and value for money. And lets be real, the US system is set up to provide revenues and profits for providers and suppliers. It’s a bit like saying Tammany Hall provided the best government services because it cost the most, when huge chunks of the money were getting diverted off into corruption.
Furthermore, it’s also a stretch for Rogoff to suggest that by definition government spending creates lower innovation compared to private spending. After all government spending led to the creation of the Internet and biotechnology. Private spending created reality TV. And despite the fact that there is no private spending on defense, well the boys and girls in the US military are no longer riding around on horses pulling gun carts. Somehow innovation and progress seems to find a way to happen even in government sponsored sectors. And if we want to drag real communists into the equation, the reason that we’re not all speaking German is that Hitler lost WWII to a nation that ten years before he invaded was inhabited by peasants. Yup, unpleasant as it might have been, Stalin’s Great Leap Forward in the 1930s was by far the fastest period of economic growth seen in any nation, probably any time…..just in time to save our arses in 1942-4.
I’m not exactly advocating purges, slave labor camps, collectivization and enforced Ten Year Plans as a panacea for the future of health care (although David Brailer keeps going on about his ten year plan). But the overall point is that greater government involvement in spending and regulation of health care doesn’t necessarily mean the disaster in services and innovation that Rogoff suggests. And there are excellent reasons from the “socialist” angle for greater government involvement in health care than we have now.
The first is the fallacy that there can be such a thing as a private health insurance market with free use of underwriting. Social insurance (or universal insurance), in which everyone pays in and everyone receives at least a basic level of benefits is the only way to get around the problem of the uninsured and the uninsurable. It of course means a relative redistribution of income from the healthy and wealthy to the poor and sick, but in fact that can be budget neutral to the healthy and wealthy if the overall price tag is kept down. That though would require a redistribution of income from the health care sector to the rest of society. Such universal insurance is good enough for everyone over-65 in this country and good enough for everyone else in the developed world, but the concept just can’t seem to get the attention of the American public enough to force it past the “special interests” in Congress. And everyone (apart from actuaries and underwriters and some participants in the system) suffers as a result.
The second is the role of government or someone like it as a clearinghouse of information or as a standards-setting body in a market where information access is very lopsided. Health care is very, very complex and someone has to provide decent information (preferably with some regulatory teeth) so that consumers/patients are not at the mercy of providers and suppliers who know far more than they do and in whom most patients still are forced to place their trust blindly. This is the role of the NICE in the UK, and in theory ought to be the role of the FDA here. Adding an economic element to that role by giving information on value for money would probably be derided as socialism by Rogoff’s “capitalists”, but is a rational role for government. And one they are likely to add as spending increases — of course the Brits and Aussies already have done so to some extent, and are linking cost-effective performance to payment.
So overall I don’t think there’s any basis for suggesting that if we have more “socialism” in health care — and by that I’m using Rogoff’s meaning of government spending, regulation and income redistribution — we will necessarily have worse services or lower innovation. Although we may have lower drug prices and a less profitable health care industry. Anyone awake during the last three months of Vioxx breast-beating is becoming painfully aware that expensive “innovation” can be costly for the wrong reasons and actually not be innovative–COX-2s didn’t really do what they were supposed to do (reduce GI problems) but they did cost a lot more than NSAIDs in both money and increased heart disease. But it’s that kind of “innovation” that Rogoff correctly says that Americans are paying more for than anyone else.
However, Rogoff is making a very important point when he discusses the likely trade-offs between basic health care and lifestyle enhancements that will dominate the politics of health care for the next century. We’ve already seen this begin with the medicalization of social afflictions (ugly teeth, small breasts), the medicalization of several “diseases” that aren’t really diseases (impotence, shyness), and the medicalization of old age (osteoporosis, prostate cancer). Now the nano-gurus are discussing the medicalization of death — which will presumably lead to a cure, or at least a delay, for it at a hell of a price.
As more and more health care services become luxury goods, there is a justifiable discussion about what’s a basic necessity and what should someone have to pay for out of their discretionary income. At the moment no-one’s seriously suggesting that your boob job or teeth whitening should be other than an individual expense, or that your cancer treatment is a luxury good to be chosen if your mood and wallet fits. But clearly the middle of that continuum will continue to fill up.
This leads me to what has been called the mocha-Frappuchino problem. I read an article once (that I can’t find anymore) that discussed the increase in productivity of the US workforce since the 1930s. It’s doubled. Which means that we could work half the time and have a 1930s standard of living, or we could work as hard as we do now and have more stuff. The author noted that in the 1930s you couldn’t get a Mocha Frappuchino; so you’ve been spending Wednesday 1pm through Friday afternoon working for your Frappuchino (or similarly frothy goods and services).
We’ve always thought of health care as an “essential”. And eventually even in the US I believe we’ll figure out a way to solve the problem of creating an equitable and sustainable social insurance model for that “essential”. But increasingly, the health care Frappuccino will be paid for and delivered privately, in a separate system. Of course it’s the blurring of those two systems that concerns bleeding heart liberals like me, as that can well lead not as it has done here for the Medicare population, as society giving Frappuccinos to everyone, but instead society deciding to take away essential services from those who can’t afford Frappuccinos.
And that will be the real socialism versus capitalism battle of the next decades.
Matthew Holt is the founder of The Health Care Blog
As I’ve always suspected, Health Care = Communism + Frappuccinos published first on https://wittooth.tumblr.com/
0 notes