Editor’s note: This guest post is by Saurabh Jha, MBBS, a radiologist at the University of Pennsylvania. This post will also be published on The Health Care Blog, where Jha is the associate editor.
A popular meme is that the US spends more on healthcare than other developed nations but has nothing to show for that spending. This is different from saying that the US spends more, but achieves something, but the something it achieves is so little that it isn’t worth the public purse. The latter is difficult to assert because the asserter must say how little is too little in regards to how much is spent, and why. It is easier believing the excess spending has no effect whatsoever, zilch in fact, because this absolves one from having to apply a value judgment on how much a life is worth. This meme, a convenient heuristic, like other convenient heuristics, is wrong.
A recent study looked at trends and outcomes in the management of abdominal aortic aneurysm (AAA) in the US and the UK. An aneurysm, dilation of the aorta, is more likely to burst the bigger it gets. Aneurysms should be repaired before they rupture because the mortality of ruptured aneurysms can be 50%. The study, which analyzed several databases that recorded surgery, size of aneurysms, and cause of death, found that Americans repair twice as many aneurysms as the Brits, and the repaired AAAs are smaller, on average, in the US. Between 2005-2012 elective AAA repair (i.e. repair of non-ruptured aneurysms) increased from 27 to 32 per 100, 000 in the UK, and from 58 to 64 per 100, 000 in the US.
Does the increased frequency of repair of AAA in the US reap benefits? It seems so. In 2012, there were twice as many ruptured aneurysms in the UK as the US, and aneurysm-related deaths were 3.5 times higher in the UK. What, if not rupture, caused the higher rate of “aneurysm-related deaths?” Hospitalization for ruptured AAA is a more reliable metric than aneurysm-related death.
Only trends, not absolute numbers, should be inferred from secondary databases. And the trend is clear: in both the UK and the US, the rates of ruptured AAA and aneurysm-related deaths have declined, while elective AAA repair has increased. The UK has reduced aneurysm-related deaths by 20 per 100, 000 by adding only 5 per 100, 000 cases of elective repair. Again, the exact numbers may not be clear but it seems that UK has picked the low-lying fruits (large aneurysms) and the US is approaching diminishing returns.
Roughly, for 32 excess electively repaired AAAs, there are 9 fewer ruptured AAAs and 25 fewer aneurysm-related deaths, per 100, 000. These figures aren’t exact but show that repairing AAA before it ruptures has a good return-on-investment and, as far as life expectancy is concerned, more the merrier. Of note, electively-repaired AAAs have the same outcome – i.e. the same complications and therapeutic effect – in the US and the UK. Neither the skill of the surgeon, nor the attentiveness of the support staff, seems meaningfully different between the two countries.
The corollary of Americans repairing more AAAs is that the size-threshold for repair of AAA in the US is smaller than the UK. The average size of repaired AAA is 5.8 cm in the US and 6.4 cm in the UK. At the time of repair the AAA, on average, is 5.3 mm smaller in the US than UK. 5.3 mm is a lot! Risk of aneurysm rupture is non-linear – the increased risk of rupture of 65-mm vs. 60-mm aneurysm is more than the increased risk of rupture of 45-mm vs. 40-mm aneurysm, even though the difference in size in the two pairs is the same. The non-linearity of rupture risk means that excess 7-cm AAAs floating around in the UK, for example, will contribute disproportionately to aneurysm-related mortality.
Clearly, the Americans are repairing aneurysms sooner than the Brits and, in many instances, aneurysms smaller than the recommended size threshold. Further, AAA is more likely to be repaired endovascularly – i.e. by a stent – in the US. Stents have lower morbidity-mortality than open repair. In the US, there are more physicians available to stent AAAs, or more willingness in physicians to stent, or both. Why is this so?
Consider an analogy. Peter drinks more alcohol than Paul because he has more alcohol in his house than Paul. But the reason Peter has more alcohol in his house than Paul is because he drinks more alcohol than Paul – he drinks more because he has more and he has more because he drinks more. The process is recursive. Americans stent more because it pays more to stent than not to stent. But crucially, the “more stenting” is not for naught. Americans are more aggressive not only with stenting AAA, but surveillance of AAAs – I can attest to that as I read CT angiograms for AAA. The “Aneurysm Surveillance Program” puts the vigilance of the Central Intelligence Agency to shame.
The study suggests that the size-threshold for repair of AAA, presently 55/ 50 mm (men/ women), should be lower. Thresholds are derived from risk vs. benefit of an intervention – the safer an intervention, the lower the threshold for intervening. Threshold for repair of AAA was derived from a randomized controlled trial (RCT) when aneurysms were repaired by open surgery. Threshold should be revised because now stents, which are safer, are mostly used. The study is an excellent example of how analysis of a secondary database can question practice derived from an outdated RCT.
The study also hints that screening for AAAs may be beneficial. However, it won’t be easy for an RCT to show a treatment effect of mass screening for AAA, even though, undoubtedly, some lives will be saved by screening. This is because the outcome, death from ruptured aneurysm, is still an uncommon occurrence, at a population level.
In summary, Americans stent more aneurysms and stent smaller aneurysms than the Brits, increasing the longevity of some people with aneurysms. There is another message in this paper. The Americans are repairing aneurysms smaller than the recommended threshold. To state this bluntly – they’re saving lives by ignoring evidence-based medicine (EBM). This is, partly, how medicine progresses – someone ignores the status quo, which is what guidelines are. To advance science you must ignore EBM. This is a paradox until you think about it.
This is a good time to deliver my annual message to both countries. Brits: if you want American outcomes, put your money where your mouth is. Americans: if you want British healthcare spending, build more graveyards. Sometimes more is more, sometimes less is more.
—Saurabh Jha
I don’t know the status quo, but about a decade ago my GP practice wrote to me to recommend screening for AAA, saying that it wasn’t available on the NHS, and directing me to a company who would do it at a modest cost. So I stumped up.