andrew_gelman_stats andrew_gelman_stats-2010 andrew_gelman_stats-2010-322 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: Someone who works in statistics in the pharmaceutical industry (but prefers to remain anonymous) sent me this update to our discussion on the differences between approvals of drugs and medical devices: The ‘substantial equivalence’ threshold is a very outdated. Basically the FDA has to follow federal law and the law is antiquated and leads to two extraordinarily different paths for device approval. You could have a very simple but first-in-kind device with an easy to understand physiological mechanism of action (e.g. the FDA approved a simple tiny stent that would relieve pressure from a glaucoma patient’s eye this summer). This device would require a standard (likely controlled) trial at the one-sided 0.025 level. Even after the trial it would likely go to a panel where outside experts (e.g.practicing & academic MDs and statisticians) hear evidence from the company and FDA and vote on its safety and efficacy. FDA would then rule, consider the panel’s vote, on whether to appro
sentIndex sentText sentNum sentScore
1 Someone who works in statistics in the pharmaceutical industry (but prefers to remain anonymous) sent me this update to our discussion on the differences between approvals of drugs and medical devices: The ‘substantial equivalence’ threshold is a very outdated. [sent-1, score-0.214]
2 Basically the FDA has to follow federal law and the law is antiquated and leads to two extraordinarily different paths for device approval. [sent-2, score-1.124]
3 You could have a very simple but first-in-kind device with an easy to understand physiological mechanism of action (e. [sent-3, score-1.028]
4 the FDA approved a simple tiny stent that would relieve pressure from a glaucoma patient’s eye this summer). [sent-5, score-0.323]
5 This device would require a standard (likely controlled) trial at the one-sided 0. [sent-6, score-0.925]
6 Even after the trial it would likely go to a panel where outside experts (e. [sent-8, score-0.366]
7 practicing & academic MDs and statisticians) hear evidence from the company and FDA and vote on its safety and efficacy. [sent-10, score-0.188]
8 FDA would then rule, consider the panel’s vote, on whether to approve this device. [sent-11, score-0.055]
9 On the other hand you could have a very complex device with uncertain physiological mechanism declared equivalent to a device approved before May 28, 1976 and it requires much less evidence. [sent-12, score-2.182]
10 And you can have a device declared similar to a device that was similar to a device that was similar to a device on the market before 1976. [sent-13, score-3.441]
11 So basically if there was one type I error in this chain, you now have a device that’s equivalent to a non-efficacious device. [sent-14, score-0.926]
12 For these no trial is required, no panel meeting is required. [sent-15, score-0.319]
13 The regulatory burden is tens of millions of dollars less expensive and we also have substantially less scientific evidence. [sent-16, score-0.447]
14 But the complexity of the device has nothing to do with which path gets taken. [sent-17, score-0.792]
15 Only it’s similarity to a device that existed before 1976. [sent-18, score-0.861]
16 You can imagine there was nothing quite like the “NanoKnife” on the market in 1976. [sent-20, score-0.165]
17 But it’s obviously very worth a company’s effort to get their new device declare substantially equivalent to an old one. [sent-21, score-1.008]
18 Otherwise they have to spend the money for a trial and risk losing that trial. [sent-22, score-0.23]
19 So this unfortunately isn’t a scientific question — we know what good science would lead us to do. [sent-25, score-0.073]
20 law which is fundamentally flawed and leads to two very different paths and scientific hurdles for device approval. [sent-28, score-1.186]
wordName wordTfidf (topN-words)
[('device', 0.745), ('fda', 0.295), ('trial', 0.18), ('physiological', 0.147), ('panel', 0.139), ('declared', 0.121), ('market', 0.118), ('approved', 0.113), ('equivalent', 0.113), ('law', 0.103), ('paths', 0.099), ('substantially', 0.093), ('mechanism', 0.09), ('leads', 0.074), ('similar', 0.074), ('company', 0.074), ('scientific', 0.073), ('relieve', 0.069), ('basically', 0.068), ('equivalence', 0.066), ('wsj', 0.066), ('vote', 0.065), ('similarity', 0.064), ('prefers', 0.06), ('regulatory', 0.059), ('pharmaceutical', 0.059), ('less', 0.057), ('declare', 0.057), ('burden', 0.056), ('devices', 0.056), ('approve', 0.055), ('tens', 0.052), ('existed', 0.052), ('uncertain', 0.051), ('losing', 0.05), ('controlled', 0.049), ('approval', 0.049), ('patient', 0.049), ('drugs', 0.049), ('safety', 0.049), ('eye', 0.048), ('tiny', 0.047), ('likely', 0.047), ('nothing', 0.047), ('anonymous', 0.047), ('flawed', 0.046), ('morning', 0.046), ('fundamentally', 0.046), ('industry', 0.046), ('simple', 0.046)]
simIndex simValue blogId blogTitle
same-blog 1 1.0000001 322 andrew gelman stats-2010-10-06-More on the differences between drugs and medical devices
Introduction: Someone who works in statistics in the pharmaceutical industry (but prefers to remain anonymous) sent me this update to our discussion on the differences between approvals of drugs and medical devices: The ‘substantial equivalence’ threshold is a very outdated. Basically the FDA has to follow federal law and the law is antiquated and leads to two extraordinarily different paths for device approval. You could have a very simple but first-in-kind device with an easy to understand physiological mechanism of action (e.g. the FDA approved a simple tiny stent that would relieve pressure from a glaucoma patient’s eye this summer). This device would require a standard (likely controlled) trial at the one-sided 0.025 level. Even after the trial it would likely go to a panel where outside experts (e.g.practicing & academic MDs and statisticians) hear evidence from the company and FDA and vote on its safety and efficacy. FDA would then rule, consider the panel’s vote, on whether to appro
2 0.4306232 314 andrew gelman stats-2010-10-03-Disconnect between drug and medical device approval
Introduction: Sanjay Kaul wrotes: By statute (“the least burdensome” pathway), the approval standard for devices by the US FDA is lower than for drugs. Before a new drug can be marketed, the sponsor must show “substantial evidence of effectiveness” as based on two or more well-controlled clinical studies (which literally means 2 trials, each with a p value of <0.05, or 1 large trial with a robust p value <0.00125). In contrast, the sponsor of a new device, especially those that are designated as high-risk (Class III) device, need only demonstrate "substantial equivalence" to an FDA-approved device via the 510(k) exemption or a "reasonable assurance of safety and effectiveness", evaluated through a pre-market approval and typically based on a single study. What does “reasonable assurance” or “substantial equivalence” imply to you as a Bayesian? These are obviously qualitative constructs, but if one were to quantify them, how would you go about addressing it? The regulatory definitions for
3 0.26008973 1065 andrew gelman stats-2011-12-17-Read this blog on Google Currents
Introduction: I’ve been told that if you download Google Currents to your iPad or Android device, you get a blog reader that beautifies the posts and makes it look more like a magazine. I don’t have a mobile phone myself but maybe those of you who do, will find this useful.
4 0.25958917 795 andrew gelman stats-2011-07-10-Aleks says this is the future of visualization
Introduction: Here . My reaction was, It’s cute how the bars move but why is this the future? Aleks replied: Integrated in the browser, works on any device, requires no software installation. Here are more examples, for maps.
5 0.19352023 2121 andrew gelman stats-2013-12-02-Should personal genetic testing be regulated? Battle of the blogroll
Introduction: On the side of less regulation is Alex Tabarrok in “Our DNA, Our Selves”: At the same time that the NSA is secretly and illegally obtaining information about Americans the FDA is making it illegal for Americans to obtain information about themselves. In a warning letter the FDA has told Anne Wojcicki, The Most Daring CEO In America, that she “must immediately discontinue” selling 23andMe’s Personal Genome Service . . . Alex clarifies: I am not offended by all regulation of genetic tests. Indeed, genetic tests are already regulated. . . . the Clinical Laboratory Improvement Amendments (CLIA) . . . requires all labs, including the labs used by 23andMe, to be inspected for quality control, record keeping and the qualifications of their personnel. . . . What the FDA wants to do is categorically different. The FDA wants to regulate genetic tests as a high-risk medical device . . . the FDA wants to judge . . . the clinical validity, whether particular identified alleles are cau
7 0.13456298 167 andrew gelman stats-2010-07-27-Why don’t more medical discoveries become cures?
8 0.10637665 1032 andrew gelman stats-2011-11-28-Does Avastin work on breast cancer? Should Medicare be paying for it?
9 0.088324584 411 andrew gelman stats-2010-11-13-Ethical concerns in medical trials
10 0.078500263 1471 andrew gelman stats-2012-08-27-Why do we never see a full decision analysis for a clinical trial?
11 0.067641854 1520 andrew gelman stats-2012-10-03-Advice that’s so eminently sensible but so difficult to follow
12 0.060793448 284 andrew gelman stats-2010-09-18-Continuing efforts to justify false “death panels” claim
13 0.058158312 2149 andrew gelman stats-2013-12-26-Statistical evidence for revised standards
14 0.052988462 2049 andrew gelman stats-2013-10-03-On house arrest for p-hacking
15 0.052395768 2053 andrew gelman stats-2013-10-06-Ideas that spread fast and slow
16 0.051044486 532 andrew gelman stats-2011-01-23-My Wall Street Journal story
17 0.048581149 1572 andrew gelman stats-2012-11-10-I don’t like this cartoon
18 0.048488803 1844 andrew gelman stats-2013-05-06-Against optimism about social science
19 0.048207972 607 andrew gelman stats-2011-03-11-Rajiv Sethi on the interpretation of prediction market data
20 0.047892354 633 andrew gelman stats-2011-03-28-“The New Tyranny: Carbon Monoxide Detectors?”
topicId topicWeight
[(0, 0.088), (1, -0.027), (2, 0.015), (3, -0.026), (4, -0.017), (5, -0.0), (6, 0.005), (7, 0.006), (8, -0.001), (9, -0.012), (10, -0.033), (11, -0.027), (12, 0.027), (13, -0.02), (14, -0.036), (15, 0.008), (16, 0.022), (17, 0.003), (18, 0.02), (19, -0.017), (20, 0.011), (21, 0.006), (22, -0.027), (23, -0.002), (24, -0.023), (25, 0.014), (26, -0.005), (27, -0.015), (28, -0.006), (29, 0.027), (30, -0.014), (31, -0.013), (32, 0.016), (33, 0.045), (34, 0.048), (35, -0.008), (36, -0.008), (37, 0.037), (38, 0.024), (39, -0.012), (40, -0.044), (41, -0.067), (42, -0.004), (43, 0.059), (44, 0.008), (45, 0.017), (46, 0.013), (47, 0.035), (48, -0.018), (49, -0.03)]
simIndex simValue blogId blogTitle
same-blog 1 0.94052839 322 andrew gelman stats-2010-10-06-More on the differences between drugs and medical devices
Introduction: Someone who works in statistics in the pharmaceutical industry (but prefers to remain anonymous) sent me this update to our discussion on the differences between approvals of drugs and medical devices: The ‘substantial equivalence’ threshold is a very outdated. Basically the FDA has to follow federal law and the law is antiquated and leads to two extraordinarily different paths for device approval. You could have a very simple but first-in-kind device with an easy to understand physiological mechanism of action (e.g. the FDA approved a simple tiny stent that would relieve pressure from a glaucoma patient’s eye this summer). This device would require a standard (likely controlled) trial at the one-sided 0.025 level. Even after the trial it would likely go to a panel where outside experts (e.g.practicing & academic MDs and statisticians) hear evidence from the company and FDA and vote on its safety and efficacy. FDA would then rule, consider the panel’s vote, on whether to appro
2 0.83137906 314 andrew gelman stats-2010-10-03-Disconnect between drug and medical device approval
Introduction: Sanjay Kaul wrotes: By statute (“the least burdensome” pathway), the approval standard for devices by the US FDA is lower than for drugs. Before a new drug can be marketed, the sponsor must show “substantial evidence of effectiveness” as based on two or more well-controlled clinical studies (which literally means 2 trials, each with a p value of <0.05, or 1 large trial with a robust p value <0.00125). In contrast, the sponsor of a new device, especially those that are designated as high-risk (Class III) device, need only demonstrate "substantial equivalence" to an FDA-approved device via the 510(k) exemption or a "reasonable assurance of safety and effectiveness", evaluated through a pre-market approval and typically based on a single study. What does “reasonable assurance” or “substantial equivalence” imply to you as a Bayesian? These are obviously qualitative constructs, but if one were to quantify them, how would you go about addressing it? The regulatory definitions for
3 0.6927619 284 andrew gelman stats-2010-09-18-Continuing efforts to justify false “death panels” claim
Introduction: Brendan Nyhan gives the story . Here’s Sarah Palin’s statement introducing the now-notorious phrase: The America I know and love is not one in which my parents or my baby with Down Syndrome will have to stand in front of Obama’s “death panel” so his bureaucrats can decide, based on a subjective judgment of their “level of productivity in society,” whether they are worthy of health care. And now Brendan: Palin’s language suggests that a “death panel” would determine whether individual patients receive care based on their “level of productivity in society.” This was — and remains — false. Denying coverage at a system level for specific treatments or drugs is not equivalent to “decid[ing], based on a subjective judgment of their ‘level of productivity in society.’” Seems like an open-and-shut case to me. The “bureaucrats” (I think Palin is referring to “government employees”) are making decisions based on studies of the drug’s effectiveness: An FDA advisory committee
4 0.68034595 67 andrew gelman stats-2010-06-03-More on that Dartmouth health care study
Introduction: Hank Aaron at the Brookings Institution, who knows a lot more about policy than I do, had some interesting comments on the recent New York Times article about problems with the Dartmouth health care atlas. which I discussed a few hours ago . Aaron writes that much of the criticism in that newspaper article was off-base, but that there are real difficulties in translating the Dartmouth results (finding little relation between spending and quality of care) to cost savings in the real world. Aaron writes: The Dartmouth research, showing huge variation in the use of various medical procedures and large variations in per patient spending under Medicare, has been a revelation and a useful one. There is no way to explain such variation on medical grounds and it is problematic. But readers, including my former colleague Orszag, have taken an oversimplistic view of what the numbers mean and what to do about them. There are three really big problems with the common interpreta
5 0.66564679 2049 andrew gelman stats-2013-10-03-On house arrest for p-hacking
Introduction: People keep pointing me to this excellent news article by David Brown, about a scientist who was convicted of data manipulation: In all, 330 patients were randomly assigned to get either interferon gamma-1b or placebo injections. Disease progression or death occurred in 46 percent of those on the drug and 52 percent of those on placebo. That was not a significant difference, statistically speaking. When only survival was considered, however, the drug looked better: 10 percent of people getting the drug died, compared with 17 percent of those on placebo. However, that difference wasn’t “statistically significant,” either. Specifically, the so-called P value — a mathematical measure of the strength of the evidence that there’s a true difference between a treatment and placebo — was 0.08. . . . Technically, the study was a bust, although the results leaned toward a benefit from interferon gamma-1b. Was there a group of patients in which the results tipped? Harkonen asked the statis
6 0.66084212 1906 andrew gelman stats-2013-06-19-“Behind a cancer-treatment firm’s rosy survival claims”
7 0.65754741 238 andrew gelman stats-2010-08-27-No radon lobby
8 0.65134263 2121 andrew gelman stats-2013-12-02-Should personal genetic testing be regulated? Battle of the blogroll
10 0.62853199 411 andrew gelman stats-2010-11-13-Ethical concerns in medical trials
11 0.61933696 791 andrew gelman stats-2011-07-08-Censoring on one end, “outliers” on the other, what can we do with the middle?
13 0.60817361 2199 andrew gelman stats-2014-02-04-Widening the goalposts in medical trials
14 0.60529006 53 andrew gelman stats-2010-05-26-Tumors, on the left, or on the right?
15 0.60258996 21 andrew gelman stats-2010-05-07-Environmentally induced cancer “grossly underestimated”? Doubtful.
17 0.59590626 1766 andrew gelman stats-2013-03-16-“Nightshifts Linked to Increased Risk for Ovarian Cancer”
18 0.59477645 765 andrew gelman stats-2011-06-14-How the ignorant idiots win, explained. Maybe.
19 0.59280664 465 andrew gelman stats-2010-12-13-$3M health care prediction challenge
20 0.59176821 1789 andrew gelman stats-2013-04-05-Elites have alcohol problems too!
topicId topicWeight
[(9, 0.019), (15, 0.037), (16, 0.072), (21, 0.024), (24, 0.149), (41, 0.015), (43, 0.021), (53, 0.041), (54, 0.212), (84, 0.025), (99, 0.21)]
simIndex simValue blogId blogTitle
same-blog 1 0.93827903 322 andrew gelman stats-2010-10-06-More on the differences between drugs and medical devices
Introduction: Someone who works in statistics in the pharmaceutical industry (but prefers to remain anonymous) sent me this update to our discussion on the differences between approvals of drugs and medical devices: The ‘substantial equivalence’ threshold is a very outdated. Basically the FDA has to follow federal law and the law is antiquated and leads to two extraordinarily different paths for device approval. You could have a very simple but first-in-kind device with an easy to understand physiological mechanism of action (e.g. the FDA approved a simple tiny stent that would relieve pressure from a glaucoma patient’s eye this summer). This device would require a standard (likely controlled) trial at the one-sided 0.025 level. Even after the trial it would likely go to a panel where outside experts (e.g.practicing & academic MDs and statisticians) hear evidence from the company and FDA and vote on its safety and efficacy. FDA would then rule, consider the panel’s vote, on whether to appro
2 0.92454863 839 andrew gelman stats-2011-08-04-To commenters who are trying to sell something
Introduction: We screen our comments. If you link to an url of the form, http://we’re-selling-you-crap.org, then you go straight into the spam folder. If you want to contribute to the discussion here, fine. Comment without the spam links. If you want to advertise, go elsewhere. It’s customary to pay for ads. We have no plans to advertise your services for free.
3 0.90782171 1938 andrew gelman stats-2013-07-14-Learning how to speak
Introduction: I’ve been trying to reduce my American accent when speaking French. I tried taping my voice and playing it back, but that didn’t help. I couldn’t actually tell that I had a strong accent by listening to myself. My own voice is just too familiar to me. Then Malecki told me about the international phonetic alphabet, which is just great. And there’s even a convenient website that translates. For example, le loup est revenu -> lə lu ε ʀəvny I stared at Malecki’s mouth while he said the phrase, and I finally understood the difference between the two different “oo” sounds. That evening at home I tried it out on the local expert and he laughed at my attempts but grudgingly admitted I was getting better. On about the 10th try, after watching him say it over and over and staring at his mouth, I was finally able to do it! I know this is going to sound stupid to all you linguistics experts out there, but I had no idea that you could figure out how to speak better by staring at s
4 0.83599806 1676 andrew gelman stats-2013-01-16-Detecting cheating in chess
Introduction: Three different people have pointed me to this post by Ken Regan on statistical evaluation of claims of cheating in chess. So I figured I have to satisfy demand and post something on this. But I have nothing to say. All these topics interest me, but I somehow had difficulty reading through the entire post. I scanned through but what I really wanted to see was some data. Show me a scatterplot, then I’ll get interested. P.S. This is meant as no disparagement of Regan or his blog. I just couldn’t quite get into this particular example.
5 0.83248007 615 andrew gelman stats-2011-03-16-Chess vs. checkers
Introduction: Mark Palko writes : Chess derives most of its complexity through differentiated pieces; with checkers the complexity comes from the interaction between pieces. The result is a series of elegant graph problems where the viable paths change with each move of your opponent. To draw an analogy with chess, imagine if moving your knight could allow your opponent’s bishop to move like a rook. Add to that the potential for traps and manipulation that come with forced capture and you have one of the most remarkable games of all time. . . . It’s not unusual to hear masters of both chess and checkers (draughts) to admit that they prefer the latter. So why does chess get all the respect? Why do you never see a criminal mastermind or a Bond villain playing in a checkers tournament? Part of the problem is that we learn the game as children so we tend to think of it as a children’s game. We focus on how simple the rules are and miss how much complexity and subtlety you can get out of those ru
6 0.82778072 358 andrew gelman stats-2010-10-20-When Kerry Met Sally: Politics and Perceptions in the Demand for Movies
7 0.81569338 1473 andrew gelman stats-2012-08-28-Turing chess run update
8 0.81418854 1889 andrew gelman stats-2013-06-08-Using trends in R-squared to measure progress in criminology??
9 0.80666012 502 andrew gelman stats-2011-01-04-Cash in, cash out graph
10 0.80449665 1578 andrew gelman stats-2012-11-15-Outta control political incorrectness
11 0.80150151 94 andrew gelman stats-2010-06-17-SAT stories
12 0.80086428 1721 andrew gelman stats-2013-02-13-A must-read paper on statistical analysis of experimental data
13 0.79920131 2121 andrew gelman stats-2013-12-02-Should personal genetic testing be regulated? Battle of the blogroll
14 0.79672432 1105 andrew gelman stats-2012-01-08-Econ debate about prices at a fancy restaurant
15 0.78710794 696 andrew gelman stats-2011-05-04-Whassup with glm()?
16 0.78610611 2271 andrew gelman stats-2014-03-28-What happened to the world we knew?
17 0.78486949 1818 andrew gelman stats-2013-04-22-Goal: Rules for Turing chess
18 0.78247529 1237 andrew gelman stats-2012-03-30-Statisticians: When We Teach, We Don’t Practice What We Preach
19 0.77805769 867 andrew gelman stats-2011-08-23-The economics of the mac? A paradox of competition
20 0.77758825 2149 andrew gelman stats-2013-12-26-Statistical evidence for revised standards