I got asked the other day to comment for Wired on the role of AI in Covid-19 detection, in particular for use with CT scanning. Since I didn’t know exactly what resources they had on the ground in China, I could only make some generic vaguely negative statements. I thought it would be worthwhile to expand on those ideas here, so I am writing two blog posts on the topic, on CT scanning for Covid-19, and on using AI on those CT scans.
As background, the pro-AI argument goes like this:
CT screening detects 97% of Covid-19, viral PCR only detects 70%!
A radiologist takes 5-10 minutes to read a CT chest scan. AI can do it in a second or two.
If you use CT for screening, there will be so many studies that radiologists will be overwhelmed.
In this first post, I will explain why CT, with or without AI, is not worthwhile for Covid-19 screening and diagnosis, and why that 97% sensitivity report is unfounded and unbelievable.
Next post, I will address the use of AI for this task specifically.
By THOMAS WILSON PhD, DrPH and VINCE KURAITIS JD, MBA
A recent study in the New England Journal of Medicine reported on the results of a “hotspotting” program created by the Camden Coalition of Healthcare Providers (Camden Coalition). Hotspotting targets interventions at all or a subset of healthcare superutilizers – the 5% of patients that account for 50% of annual healthcare spending.
of the study were disappointing. While utilization (hospital readmissions)
declined for the hotspotting group, the declines were almost identical in the
control group. At least three headlines
implied that the conclusion of the study was that hotspotting care management
approaches have been proven not to work:
explain, we believe that much of what’s going on here can be explained by one
or both of what we call “RTM Traps” (regression to the mean traps).
essay, we will:
Define RTM (regression to the mean)
Explain the RTM Traps and how many
have fallen into the traps
Suggest how to avoid the RTM Traps
our POV is relevant to clinical, technical, and executive staff in the many
organizations focusing on the superutilizer population – hospitals, physicians,
ACOs, health plans, community groups, etc.
By JOHN JAMES, ROBERT R. SCULLY, CASEY QUINLAN, BILL ADAMS, HELEN HASKELL, and POPPY ARFORD
Political forces trying to shape and reshape American healthcare without hearing the voice of patients provided the rationale for this work. Our experiences as patients, caregivers, and users of media sources cause us to worry. The Patient Council of the Right Care Alliance developed 6 questions to form a national survey of Americans to guide policy makers. The questions and our rationale were as follows:
3) I will get an infection while receiving treatment. Healthcare-associated infections have dropped somewhat in the past decade, yet there are still about 720,000 infections and 75,000 deaths per year from healthcare-associated infections. Many of these are becoming nearly impossible to effectively treat. The improper use of ordinary antibiotics continues to be a problem in clinical settings.
Sharing a hotel room, however, does not a marriage make. In order to get better digital health interventions to market faster, we need what I’m calling a Partnership for Innovators, Policymakers and Evidence-generators (PIPE). As someone who functions variously in the policy, tech and academic worlds, I believe PIPE needn’t be a dream.
It is easy for armchair activists to bash randomized controlled trials (RCTs) with clever methodological critiques. However, it takes a lot of effort and coordination to pull off an RCT successfully. In this episode of Radiology Firing Line, I speak with Dr. Mark Neuman and Lakisha Gaskins, principal investigator and research project manager of the REGAIN trial, respectively, about the logic, challenges and intricacies of conducting an RCT. The Regional versus General Anesthesia for Promoting Independence After Hip Surgery (REGAIN) trial is an ongoing pragmatic, multi-center RCT, funded by PCORI, which randomizes patients with hip fractures to regional or general anesthesia.
Guests: Mark Neuman MD MSc, is an Associate Professor of Anesthesiology and Critical Care at the University of Pennsylvania. He is a senior fellow at the Leonard Davis Institute of Health Economics. He’s a former RWJ Scholar. Lakisha Gaskins is a research coordinator with extensive experience recruiting patients for RCTs.
Listen to our conversation on Radiology Firing Line Podcast here.
Saurabh Jha is a contributing editor to THCB and host of Radiology Firing Line Podcast of the Journal of American College of Radiology, sponsored by Healthcare Administrative Partner.
September was an important month in oncology—especially for lung cancer. The World Conference in Lung Cancer (WCLC) 2018 gave us some important practice-changing results, also leading to four NEJM publications. The trial with most public health impact is unfortunately not published yet. It’s the NELSON trial that randomised more than 15000 asymptomatic people at high risk of lung cancer to either CT-based screening for lung cancer or to no screening and found a significant reduction in lung cancer mortality rates among the screened cohort compared with the control cohort. This reduction was more pronounced among women, although they constituted only 16% of the trial population. I am looking forward to reading the full publication and am particularly interested in knowing if there were any differences in all-cause mortality rates and the rates of overdiagnoses.
A new ALK-inhibitor on the block—brigatinib—has significantly improved PFS versus crizotinib when used as first-line therapy in ALK-positive non-small cell lung cancer (NSCLC) patients. However, I assume that it will be difficult for brigatinib to replace alectinib in this setting, since the latter has already been tested in two different RCTs and has more mature data.
With Keynote 407, pembrolizumab has entered into the treatment arsenal for squamous NSCLC by improving overall survival in combination with chemotherapy versus chemotherapy alone as a first-line regimen. However, when A B is compared with A, it is important to know whether A B is better than A followed by B. In this trial, 32% of patients who were in the control arm received a PD-1 inhibitor upon progression. Nivolumab is already approved as a second-line option in this setting after first-line chemo; so how much benefit in Keynote 407 is due to more than half of control arm patients not getting PD-1 inhibitor at all versus the benefit of combining pembrolizumab with chemo upfront is an important question.
This month, we saw historic turnout at the polls for midterm elections with over 114 million ballots cast. One noteworthy observation regarding voter turnout is record rates of participation by younger voters aged between 18 to 29 years old. Around 31 percent of people aged 18 to 29 voted in the midterms this year, an increase from 21 percent in 2014, according to a day-after exit poll by Tufts University.
Surely their political engagement counters the criticism that millennials are disengaged and disconnected with society and demonstrates that millennials are fully engaged when issues are relevant to them, their friends, and their families. Why, then, do we not see the same level of passion, engagement and commitment when young adults are asked to consider their health and well-being?
I have had the privilege of being a member of the National Heart, Lung and Blood Institute-funded Coronary Artery Risk Development in Young Adults (CARDIA) study research team. In over 5,000 black and white adults who were initially enrolled when they were 18 to 30 years old and have now been followed for nearly 35 years, we have described the decades-long process by which heart disease develops. We were able to do this because, in the 1980s when these studies began, young adults could be reached at their home telephone numbers. When a university researcher called claiming to be funded by the government, there was a greater degree of trust.
Unfortunately, that openness and that trust has eroded, particularly in younger adults and those who may feel marginalized from our society for any number of valid reasons. However, the results—unanswered phone calls from researchers, no-shows at the research clinic and the absence of an entire group of adults today from research studies, looks like disengagement. Disengagement is a very real public health crisis with consequences that are as dire as any political crisis. Continue reading…