 The lovely Harriet Hall is next. She has a skeptical look at screening tests. Yeah, sure. It's the Skeptok. Yes, the Skeptok. She writes for Skeptok Magazine, Skeptok Enquirer. She's a contributing editor to science-based medicine, of course. Her haiku, because yeah, a skeptical look at screening tests is the talk. So this is, screening tests can tell if you at some point ever took a screening test. Please welcome Harriet Hall. That's the first. I've never been called a sorbet before. I want to start by apologizing for my worst voice. I lost my voice a couple of days ago, and I still have a frog in my throat. It's better to sound like a frog than to sound like a duck. Screening tests. We've all had them. We're all likely to have more of them in the future. There's a lot of misunderstanding and misinformation about what screening tests can and can't do. And I'm going to try to elucidate some of the issues. In the first place, what is a screening test? Well, if you do mammograms on a whole population of women that don't have any symptoms, that's a screening mammogram. But if a woman has a lump in her breast and goes to her doctor and he orders a mammogram, that's a diagnostic mammogram. Same tests, different circumstances. There are a lot of tests that are excellent for using as a diagnostic test, but they're not good for screening tests at all. Now, we're constantly being admonished to get tested for one thing or another. Dear Abby got a letter from a woman who was screened for kidney disease and found that she had a mild decrease in kidney function. And Abby was shocked to learn that 26 million Americans have chronic kidney disease. And she advised all of her readers to get their kidneys checked. That was terrible advice. Shame on you, Abby. It seemed like good advice because, gee, if you had something wrong with your kidneys, you'd want to know about it, right? But by the same token, if you had anything wrong anywhere in your body, you'd want to know about it. So it might seem that the best thing to do would be to test everybody for everything. And that would be stupid because you'd get all kinds of false positives. You'd pick up a lot of minor harmless anomalies that never would have caused any problem. They're expensive. And screening tests can hurt people. They can do more harm than good. Here's one example. You could do urine cultures as a screening test. And you'd find that 30% of elderly women had asymptomatic urinary tract infections. But treating those infections offers no benefit whatsoever. And if you treat them with antibiotics, you're risking side effects and you're helping develop antibiotic resistance. So you want to be selective about what you've picked for screening tests. And here are some of the criteria that they've come up with. The disease should have serious consequences. The screening population should have a high prevalence of the condition. I mean, you wouldn't want to screen six months old babies for gonorrhea, for example. There shouldn't be too many false positive or false negative tests. The test should, to detect disease before the critical point, the test should be safe, affordable, available. And most importantly, there should be a treatment. There's no point in picking up a disease if you can't do anything about it. And it should be a situation where treating early in the disease has an advantage over treating later on after the symptoms of the period. So would this be a good screening test? The press release said breaking news, new blood test, 87% accurate in detecting Alzheimer's with an 85% specificity rate. Now we can diagnose it earlier so it can be treated more effectively. Well, I'm sorry to say there is no effective treatment for Alzheimer's. So just on that basis alone, it wouldn't be a good screening test. But let's forget that for a minute and look at the numbers. What does it mean to say a test is 87% accurate and 85% specific? Well, it means the test is positive in 87% of people who have Alzheimer's. That's called the sensitivity. The test is negative in 85% of those who don't have Alzheimer's. That's the specificity. But that alone doesn't tell us anything because it makes a big difference how prevalent the condition is in the population that you're going to screen. So let's run through an example. What if you had a population of 10,000 people where the prevalence of Alzheimer's was 5%. Well, that would mean that in that population 500 people had Alzheimer's and 9,500 didn't. Of the 500 who have Alzheimer's, 87% of those, 435 people would have a positive test. But 65 of them would have a false negative test. And they'd be led to think that they didn't have Alzheimer's when they really did. Of the 9,500 people who don't have Alzheimer's, 85% of those, 8,075 people would have a true negative test. But 1,425 would have a false positive test. And they'd be led to think they had the disease when they really didn't. So what you really want to know is if you test positive, what's the likelihood that you actually have the disease? To find that out, you take the number of true positives and divide by the total number of positives. That comes out to 24%. We call that the positive predictive value. If you test negative, what's the likelihood that you really don't have the disease? Well, in this case, it's 99%. And that's the negative predictive value. So this test could be great for ruling out Alzheimer's, but it wouldn't be any good for ruling it in. Yes, I did have my mammogram today. Why do you ask? Well, here's a woman who's having her breast squished flat for a mammogram. In a few minutes, she's going to go get dressed and go home. And in a few days, she's going to get a pulse card in the mail. The pulse card will say either your mammogram was normal or it will say something was suspicious. We want you to come back for more tests. Now, if she gets that second pulse card, just how worried should she be? What's the likelihood that she actually has cancer? Well, I've put some percentages up on the left-hand side of the screen. And I want to have a little audience participation exercise here. I want you to pick one of those percentages. Please don't vote if you saw this in the workshop the other day. But pick one and stick to it and don't be influenced by other people raising their hands. Try to be honest. How many of you think that the probability that she actually has cancer is close to 100%? How about 90%? 70%? 50%? 30%? 10%? This is just looking at the general population of everybody. Okay, the answer is 10%. And if you guess wrong, don't feel bad because they did a study in Europe where they asked doctors. And most of the doctors got it wrong. And most of them guessed something around 90%. The percentage is going to be better if she's in a high-risk group, if she's older or has a family history or has other risk factors for breast cancer. But the important point is not the number, but the idea that a positive screening test is not a diagnosis. The most screening tests you need to do for their testing. Now the license plate says mammograms save lives. And they do. But just how many lives do they save? We tend to overestimate the benefit of mammography for three reasons. Lead time bias, length bias, and overdiagnosis bias. With lead time bias, the arrow represents the development of cancer in a woman who dies at age 60. If she waits until she has a lump in her breast at age 50 and then gets diagnosed, the time from diagnosis to death is 10 years. If she has a mammogram at age 45, then the time to death is 15 years. So it looks like mammography has increased the time to diagnosis, the time to death, but it hasn't really changed anything. It's just an artifact. The time from diagnosis has increased, but the survival rate has not increased. Length bias is because some cancers are slow growing. And they're around for a longer period of time. So if you screen for cancers at any one point in time, there's going to be an overrepresentation of the cases that are slowly progressive. So it will make it look like the survival time has increased, but it really hasn't. And then screening detects cancers that spontaneously resolve. Cancers that never would have progressed or cancers that progressed so slowly that it never would have made a difference. And we know this because women who have been screened have 22% more diagnoses of invasive cancer. And screening can't cause invasive cancer, so obviously there were invasive cancers in the other group that spontaneously resolved. The mammogram recommendations have changed. They used to be every year starting at age 40. Now they're every two years starting at age 50. And mammography isn't as good as most people think it is. If you screen 1000 women for 10 years starting at age 50, 10 of those women will be overdiagnosed and treated unnecessarily. Up to 15 of them will be diagnosed earlier without any effect on the final outcome. 500, half of them will have one false alarm during those 10 years. And half of those 250 will undergo biopsy. With all that, one life will be saved. 999 of those women would have lived just as long if they'd never had a mammogram. Now, again, these are for women in the whole population. And if you look at women that are at higher risk, the benefits of mammography are considerably greater. Now, I started with mammograms because it's polite to let ladies go first. And now for the boys. My mommy says you got one of those and we got brains. Ha ha! Boys are different under the diapers. And part of that equipment is a prostate gland. And prostate glands tend to develop cancer the more as you get older. But it isn't always harmful. A prostate cancer is found in 80% of autopsies where men died of something else. So a lot of more men die with prostate cancer than because of it. Now, the PSA test, the prostate screening antigen test, is a blood test that they use to detect prostate cancer. It is not a yes or no test. You have to consider the patient's age and risk factors and how fast the PSA level is rising. And you have to pick a cutoff level, which is really arbitrary. Now, if the PSA test is positive, then the next step is to do biopsies. This is usually done by needle biopsies. And they'll take 12 samples, six on each side. And that'll pick up cancer in about 25% of these people. But if you go back and you repeat biopsies on the negative ones, you'll find cancer in another 25% to 30% more. And theoretically, if you could examine every single cell in the prostate, you'd probably find a cancer cell or two in almost everyone. So you've got to decide when to stop biopsy. If the biopsy finds cancer, the next step is treatment. And the treatment for prostate cancer is not benign. In a large study in Europe, they found that screening with PSA tests reduced the number of deaths from prostate cancer by seven for every 10,000 men screen. You can look at the number needed to screen and the number needed to treat. You'd have to screen 1,068 men and treat 48 men to prevent one death from prostate cancer. But here's the kicker. There was no reduction in all-cause mortality. The overall death rate was exactly the same for men who got screened and men who didn't. If you have a low-grade localized prostate cancer, you have the option of watchful waiting and not doing anything until there are signs that it's spreading. And they did a study where they compared surgery to observation. And they found that surgery didn't reduce the number of deaths. And there were complications with surgery in the first 30 days, one-fifth of patients had complications including deaths. Two years later, 17% of the surgery group were incontinent, 81% had erectile dysfunction, and 12% of them had bowel dysfunction. The media, the popular advice is get tested, it could save your life. The expert advice says don't get tested, it does more harm than good, mainly through impotence and incontinence. The United States Preventive Services Task Force is an independent group of experts that keeps correct with all of the latest literature and updates recommendations that are available online. They're not perfect, but they're the most trustworthy source of recommendations there is. And I rely on them very much. It was never meant to be a cookbook, and they save themselves on the website that they recognize that skilled clinicians serve their patients by individualizing recommendations to the specific circumstances, values, and perspectives of the individual patient. So it's a starting point. Now you hear a lot of emotional anecdotes. One doctor wrote an article that he entitled, The New York Times Killed My Patient, because his patient had read an article in The New York Times saying don't get PSA tested, so he didn't get tested, and sure enough he developed invasive prostate cancer and died. But another doctor wrote an article about one of his patients who insisted on a PSA test, and it resulted in a diagnosis of low grade localized cancer, and he had the option of watchful waiting. But he couldn't stand the idea of living knowing that there was a cancer in his body. He was afraid of surgery, so he opted for radiation. He developed rectal pain and bleeding that lasted for years. He became impotent. He lost bladder control, and he told his doctor, I'd rather be dead than live wearing adult diapers for the rest of my life. Now when the USPSTF recommendations first came out, the American Urological Association disagreed with them. They thought that testing was more valuable and should be done on more people. But they've changed their thinking and they've just recently come out with new recommendations, and now they say definitively do not screen men under 40 or over 70, men with a life expectancy of less than 10 to 15 years, men between the ages of 40 and 50 who are at average risk, and it says consider screening, not do screening, but consider it for men 65 to 65 to 69 who are at average risk or high risk men of any age. And before testing, they recommend that the doctor and the patient discuss the risks and benefits and make a joint decision. Now here are some of the screening tests that they recommend for everyone. Blood pressure testing and testing for obesity, checking people's weight. Some tests are recommended only for certain age groups or risk groups. A PAPS mirror now is recommended between the ages of 21 and 65, every three to five years. Colorectal cancer screening only from age 50 to 75. Here's one that surprised me. They recommended that you only screen for diabetes in patients who have high blood pressure. And they recommended that cholesterol be screening be done in men over 35 in women, but in women and younger men only if they're at increased risk. So the guidelines are not one size fits all. They vary by age group and sex. They're different for high risk groups. They're different for pregnant women. And sometimes they're complicated. Screening for osteoporosis is recommended for women age 65 years or older. And for younger women, his fracture risk is equal to or greater than that of a 65-year-old white woman who has no additional risk factors and men should not be screened. Well, when it gets that complicated, you almost need a translator to tell you where that applies to you. That's what your doctor is for. The USPSTF has said definitely do not do these tests as screening tests because they do more harm than good. The annual chest x-ray, the TB time test, the scoliosis check, the EKG, and teaching patients to examine their own breaths and testicles. Sometimes they don't know and they say so. For glaucoma they say we have insufficient evidence to recommend for or against screening. We've recently found out that routine physicals don't do anything to reduce mortality. That can still be worthwhile to visit the doctor routinely, but the actual physical exam, the laying on of hands and looking in the ears and listening to the chest with the stethoscope, they're a waste of time in someone who doesn't have any symptoms. They become very valuable when symptoms have developed. Our theme is fakers this year, so I want to talk about some fakers. This young man went to the base exchange and just bought one of every ribbon they had in stock, and he's not fooling anybody. Ultrasound screening, you may have seen ads for these people. There's several companies that do it. They come to town, they set up in a YMCA or local church, and they offer all kinds of tests, EKGs, echocardiograms. They do ultrasounds on your carotid arteries. They screen for abdominal erotic aneurysms and for circulation problems in the legs. They all come with big testimonials saying, I got screened and it saved my life. I would have had a stroke if I hadn't had these screening tests, and they make a big deal of, we'll do tests on you that your doctor wouldn't order, but that's very deceptive because your doctor will order these tests for a very good reason. They do more harm than good. The USPSTF recommends against screening for almost everything on their list. The one thing that they recommend screening for is abdominal erotic aneurysm screening, AAA. But they recommend that only be done once and only in men aged 65 to 75 who have ever smoked. But these screening companies will do it on everybody who walks in the door, and they'll do it next year when you come back so they can have more of your money. You may have heard of electrodermal testing. This is where they hook a biofeedback machine up to a computer, and they diagnose all kinds of bogus imbalances and health conditions, and then the computer tells the operator which homeopathic remedies and supplements she ought to sell you. Totally bogus. Your chiropractor may do a surface electromyogram on you to locate your subluxations. This is totally bogus, especially since there's no such thing as a chiropractic subluxation. A few years ago you may have seen ads for whole body CT scans. They had freestanding centers that would do these at the patient's request, and you don't see those around anymore. Fortunately, the FAD has gone out of fashion because one of these whole body scans exposed patients to the equivalent of the radiation from 923 chest x-rays, and it found abnormalities in up to 86% of asymptomatic people. Those people then would go to their doctor, and the doctor would have to try to find out if those were really significant abnormalities or not. A lot of times they'd end up doing invasive, dangerous tests, and most of the time they found nothing. Of course the patients would worry about things that they really had no need to worry about. In the UK, the pharmacies do blood sugar and cholesterol tests on people. When they started doing that, it increased the number of doctor visits, but only 10% of those people needed treatment. There are tests available today that just offer too much information. There's a company called Talking20 that has you send in a drop of your blood on a piece of filter paper, and they do 17 different tests on everybody whether you need it or not. They're planning to increase the number of tests to up to the hundreds, or even thousands of tests on one drop of blood. The tests are not validated, by the way, but even if they worry nobody needs that many tests. That's just silly. There are a lot of companies offering genomic testing, genetic testing, direct to the consumer. And they're not screening your whole genome. What they're doing is looking for specific SNPs, single nucleotide polymorphisms. And the kind of information they give you will be something like, people with your SNP have a 30% increased risk of Parkinson's disease compared to patients with other SNPs. Well, so what? There's nothing you can do to prevent Parkinson's disease, and just because you have a gene for a disease doesn't mean you're going to develop it. Genetics is not destiny. Gene expression depends on a lot of other things. It depends on environmental factors and epigenetic factors, and it depends on the interactions and effects of other genes. And there are companies that offer genetic testing with the promise that they will then give you personalized diet advice and supplements, tell you just which supplements you need to take. Well, they absolutely cannot do that yet. They're promising far more than they can do. And that kind of testing is really quackery. Unfortunately, our access to genetic information currently exceeds our understanding of what that information actually means. It's just not ready for prime time yet, folks. The latest wrinkle is the Ubiome Kid, where they will test the genome of all of the bacteria that live in you and on you. And I don't know what you're supposed to do with that information. So as we fight the fakers, we're not just fighting charlatans, we're also fighting misinformation, alternative medicine myths, overzealous disease mongering, and quackidemic medicine, where the professors in our medical schools, the very lot to know better, are doing what this Carol and the cartoon is doing, turning to the duck side. That's all. This is Kent Duck, Harry Hall.