Should you be screened for lung cancer? Maybe not, and here's why

If you are a lifelong heavy smoker age 55 years or older, the U.S. Preventive Services Task Force believes that screening for lung cancer with CT scans may save your life. Today the Task Force released provisional recommendations that assigned a "B" letter grade to this preventive service, which, if eventually finalized, would place CT scans alongside established cancer screening tests such as mammograms and Pap smears and mandate that health insurers cover these scans without cost sharing for eligible patients.

The scientific basis of the USPSTF's new recommendations is a large U.S. study sponsored by the National Cancer Institute that concluded in 2011 that heavy smokers randomly assigned to screening with CT scans had a 20 percent lower relative risk of death than heavy smokers randomly assigned to screening with chest x-rays. However, a more recent analysis by the same researchers found that the absolute benefit of CT screening in this study varied greatly according to a patient's baseline risk of lung cancer death. For the highest-risk patients, the chance of CT scans preventing death from lung cancer is about 1 in 100 (a 5% risk of lung cancer death over 6 years without screening minus a 4% risk with screening); for lower-risk patients, that chance is about 1 in 5000 (a 0.1% risk of lung cancer death without screening minus a 0.08% risk of death with screening).

For patients in the lower-risk group, for whom the potential lifesaving benefits of CT scans are very small, the downsides of the screening test become considerably more important. Screening tests have harms just like any other medical procedure, and it's important for your doctor to thoroughly review those harms with you if you are considering screening. In my mind, there are at least 4 good reasons for current or former smokers to think twice about being screened for lung cancer:

1. The risk of developing cancer from the CT scan itself isn't trivial. An analysis published in the Archives of Internal Medicine found that a typical chest CT scan exposes patients to the radiation equivalent of more than 100 chest X-rays, and that at age 60, an estimated 1 in 1000 women or 1 in 2000 men would eventually develop cancer from that single scan. Although some imaging centers now use lower radiation doses, repeating these lower-dose CT scans annually still adds up. (It hasn't been long enough since the conclusion of the NCI's lung cancer screening study to measure how much these scans increased the participants' risk for other cancers.)

2. False alarms are extremely common. In the NCI's study, more than 96 percent of all positive results turned out to be false positives, and in a previous CT screening study, 1 in 3 patients had at least one false-positive result after undergoing only two CT scans. Of those patients, 1 in 14 needed an invasive lung biopsy to be sure they were cancer-free. Such diagnostic procedures for lung cancer can themselves be life-threatening: in 2005, former Canadian prime minister Brian Mulroney (a longtime heavy smoker) spent several weeks in the intensive care unit after postoperative complications from surgery to remove two lung nodules found on a screening CT scan that turned out to be non-cancerous.

3. A CT scan for lung cancer could find some other unrelated abnormality that will require further investigation; in the NCI's study, this occurred in about 1 in every 13 patients. You might think this is a good thing, but most of these abnormalities (known as "incidentalomas") turn out to be false alarms, too. In fact, in 2008 this very same Task Force decided against endorsing CT screening for colorectal cancer due to concerns that invasive testing to definitively diagnose all of the abnormalities that CT scans turn up could easily outweigh the cancer-prevention benefits.

4. Finally, even if screening catches a true lung cancer early, there's no guarantee your prognosis will be better. This is due to "overdiagnosis," or the unnecessary diagnosis of a condition (typically cancer) that will never cause symptoms in a patient's lifetime, either because it's so slow-growing or the patient dies from some other cause. (Statistics show that most lifelong smokers will die from heart disease, not lung cancer.) An estimated 1 in 3 breast cancers detected by screening mammograms is overdiagnosed, and a 2007 study published in the journal Radiology suggested that the proportion of lung cancers overdiagnosed by CT scans could be as high or higher, especially in women. But because there's no way of knowing at the time of diagnosis if a lung cancer will be fatal, inevitably most of these patients will be needlessly subjected to the side effects of treatment - making the "cure" worse than the disease.

I expect that some of my patients will decide to be screened with CT scans for the obvious upside emphasized by the U.S. Preventive Services Task Force: a small chance of preventing death from lung cancer. But before they make this decision, I will counsel them to carefully consider the more likely downsides, and only choose testing if they're prepared for these too.

**

This post's title is modeled after the title of the excellent book Should I Be Tested for Cancer? Maybe Not and Here's Why by H. Gilbert Welch, MD, MPH. Portions of this post are revised and updated from a previous post on Common Sense Family Doctor.

Conservative Medicine: the big idea (and title) of my book

A few days ago I was reminiscing about my family medicine residency, where at one of my outpatient training sites I provided care to Amish patients in rural southern Lancaster County, Pennsylvania. The Amish didn't believe in checkups and rarely visited doctors, so we knew that when they did come in, there was usually a serious problem. Abdominal pain was more likely to be from perforated appendicitis than indigestion; leg pain was more likely to be from a life-threatening traumatic injury than a sprained ankle; and so on. There was nothing routine about providing health care for this population, and I found the experience refreshing.

On the other hand, my residency training also made me a big believer in preventive medicine, which is why I subsequently spent several years working with the U.S. Preventive Services Task Force, searching the literature for evidence to distinguish effective screening tests (such as screening for osteoporosis in older women) from ineffective and potentially harmful ones (such as screening for prostate cancer in older men). Unfortunately, as in other areas of health care, preventive medicine has been corrupted by big business, fear of lawsuits, and the widely held (and generally wrong) notion that more care is better care. As a result, unscrupulous companies such as Life Line Screening take advantage of vulnerable elders by selling packages of unnecessary tests at health fairs and churches; obstetricians continue to perform annual pelvic examinations even though these exams have no value in nonpregnant women without symptoms; and gastroenterologists tell patients with normal screening colonoscopies to return for repeat screenings much sooner than the appropriate interval of 10 years.

Looking back on my nearly 4 years of blogging, an overarching theme has been finding the "sweet spot" between too little care ("Amish medicine") and excessive medical interventions ("Life Line medicine"). This sweet spot is the starting point for a book that I'm planning to write with the working title of Conservative Medicine. It would have been easy to borrow the title of this blog and just call it "Common Sense Family Doctor," but the reality is that a lot of truths about preventive care and screening aren't common sense, and family doctors are not the only health professionals with expertise on these subjects.

The big idea of Conservative Medicine is that much of what passes for preventive health care in this country (and will be paid for by the implementation of the Affordable Care Act, aka Obamacare) is useless, unhelpful, and likely to make patients less, not more, healthy. There is no such thing as a harmless test; seemingly "routine" tests may do more harm than good; and even the well-established preventive health physical has little evidence of benefit.

My next several posts will flesh out Conservative Medicine by addressing the 5 questions that I'm told every prospective author needs to answer in a serious nonfiction book proposal:

1. What's the book about?
2. What's new about the book?
3. Why are you the best person to write it?
4. Why is now the right time for it?
5. Who needs to read it?

This book proposal is a work in progress, and this set of posts will in some ways be a departure from the usual fare of Common Sense Family Doctor. For one thing, I hope to receive many constructive comments about how to best shape my book and distinguish it from similar books (for example, Shannon Brownlee's Overtreated, H. Gilbert Welch's Overdiagnosed, and Otis Brawley's How We Do Harm). I want my book to be about a lot more than prostate cancer politics, but welcome suggestions about how to incorporate my compelling personal narrative into the larger story of a health system that has largely lost sight of what health really means. To my longtime and occasional readers, thanks in advance for your words of wisdom.

Screening-illiterate physicians may do more harm than good

On the first day of the clinical preventive medicine course that I teach every spring, I review the concept of lead-time bias and its potential to make a screening test look more effective than it really is (or, effective when it's not). In a previous blog post, I explained how lead-time bias deceptively improves 5-year survival statistics. The basic idea is that by advancing the time in the disease course that cancer (or some other condition) is detected, screening will always increase the percentage of patients who survive for 5 years or more, even if it doesn't do anything to reduce mortality. This concept is as basic to the appropriate use of screening tests as vital signs are to the practice of medicine. In my opinion, any physicians who don't understand lead-time bias ought to have their test-ordering privileges suspended until they do.

Unfortunately, a study published last year in the Annals of Internal Medicine concluded that a whole lot of clinicians require remedial education regarding lead-time bias. A national sample of more than 400 primary care physicians were provided scenarios about the effects of two hypothetical screening tests. The first test improved 5-year survival from 68 to 99 percent, and the second reduced the mortality rate from 2 deaths per 1000 to 1.6 deaths per 1000. 95 percent of surveyed physicians said that they would "definitely" or "probably" recommend the test that improved 5-year survival, even though this information (which is based on lead-time statistics associated with screening for prostate cancer) provides absolutely no evidence that the test improves patient outcomes. In contrast, considerably fewer physicians were enthusiastic about the test that actually lowered the mortality rate, perhaps because the absolute risk reduction seemed unimpressive by comparison.

Another disappointing finding was that almost half of surveyed physicians believed that a screening test "saves lives" if more cancers are detected in screened than in unscreened populations. The truth is, finding more cancers is a poor assurance of better outcomes. For example, a randomized trial of screening for ovarian cancer found no difference in mortality rates between women assigned to annual screening versus those receiving usual care despite 21% more cancers being detected in the screening group. This study confirmed what most medical organizations had suspected for years in recommending against ovarian cancer screening in asymptomatic women. Unfortunately, another survey found that one-third of a nationally representative sample of family physicians, general internists, and obstetricians nonetheless believe that ovarian cancer screening is effective. (Similar levels of ignorance exist among subspecialists, which may explain why so many urologists continue to believe that PSA screening saves lives.)

The Institute of Medicine has identified low levels of health literacy in the general population as a major obstacle to ensuring optimal health and quality of care. But how can physicians expect our patients to make informed decisions regarding screening tests when large numbers of us are functionally illiterate regarding basic screening concepts? As a medical educator, I took home this message from these studies: medical schools, residency programs, and certifying boards must devote more time and effort to improving physicians' literacy regarding screening, lest misleading survival statistics continue to fuel overuse of ineffective tests and expose countless patients to potential harm.

**

A slightly different version of the above post first appeared on Common Sense Family Doctor on March 5, 2012.

Uncertain benefits of medicating mild hypertension

In the July 1st issue of American Family Physician, Dr. Janelle Guirguis-Blake commented on a Cochrane Review that found no benefits from pharmacotherapy for mild hypertension (systolic blood pressure of 140 to 159 mm Hg and/or diastolic blood pressure of 90 to 99 mm Hg) on cardiovascular outcomes or mortality. However, the randomized trials' relatively small number of participants (fewer than 9000) and short follow-up periods (five years or less) left open the possibility that a significant benefit could still exist. Therefore, Dr. Guirguis-Blake concluded: "Larger double-blinded RCTs in this population of patients with stage 1 hypertension are needed to clarify the potential long-term benefits of pharmacologic therapy."

When existing research does not adequately answer an important clinical question - in this case, are medications superior to lifestyle modifications or no treatment for mild hypertension? - researchers invariably recommend collecting more evidence. But is performing a large randomized trial of mild hypertension management feasible, given that the standard of care set in 2003 by the Seventh Report of the Joint National Committee (JNC-7) (and reflected in this AFP Point-of-Care Guide) is to routinely identify and treat blood pressures in this range? The U.S. Preventive Services Task Force apparently thinks so; after previously declaring that the benefits of screening were "well established," the USPSTF has released an extensive draft research plan to reevaluate benefits, harms, best methods, and recommended intervals for screening for high blood pressure in adults.

With the next USPSTF statement at least a few years down the road, current evidence-based guidance on hypertension management is limited. The U.S. National Heart, Lung, and Blood Institute, which convened the previous JNC panels, recently announced in a cardiology journal its intention to stop producing guidelines. Instead, it says it will partner with outside medical groups to release its long-delayed JNC-8 hypertension guideline. Since guidelines sponsored by subspecialty societies are less likely to adhere to Institute of Medicine standards for producing unbiased guidelines, family physicians and other primary care clinicians should advocate for their organizations to participate in this process.

**

The above post was first published on the AFP Community Blog.

What pregnancy care costs, and related writings

Elisabeth Rosenthal's New York Times story on the staggering costs of maternity care in the U.S. featured a short interview with my wife, a family physician who last year gave birth to our third child with a nurse-midwife 12 minutes after arriving at the hospital. She had no postpartum complications and stayed only one night, but nonetheless received a bill of more than $6000 (we received a separate bill for newborn care). Here's what she had to say about the disconnect between this price and her delivery experience:

“Most insurance companies wouldn’t blink at my bill, but it was absurd — it was the least medical delivery in history,” said Dr. Duane, who is taking a break from practice to stay home with her children. “There were no meds. I had no anesthesia. He was never in the nursery. I even brought my own heating pad. I tried to get an explanation, but there were items like ‘maternity supplies.’ What was that? A diaper?”

Pregnancy care costs illustrate 4 reasons that Americans spend far more on health care than residents of any other country in the world: prices for services are too high; you can't find out the prices before you buy the services; the payment model incentivizes providing services that are unnecessary, nonbeneficial, or potentially harmful (e.g., extra fetal ultrasound scans, electronic fetal monitoring during labor); patients with "good insurance" (like the type that paid for the births of our two older children) are insulated from the true cost of their care. (The hospital bill for our second child's birth was more than $8500, but since we only paid $100 out-of-pocket, it was hard to get too worked up about that.)

The solutions to these problems are not simple. Unfortunately, health reform only requires that insurers provide some maternity coverage but doesn't contain skyrocketing prices or discourage doctors from doing things to pregnant women that aren't medically necessary. But how do you know what's medically necessary? Is attempting a vaginal birth after a C-section safe? How about a planned home birth? Here are answers to these and other women's health issues I've blogged about in the past:

1) The decline of VBAC: hearing hoofbeats, thinking zebras

A trial of vaginal birth after Cesarean (VBAC) is successful more than 75 percent of the time. It's a real shame that women in the U.S. are discouraged from attempting them more often.

2) VBAC and home birth: evaluating the evidence

The available evidence indicates that planned home birth is no riskier for babies, compared to planned hospital birth, than is attempting VBAC compared to choosing a repeat Cesarean delivery.

3) Unintended consequences of "pregnancy prevention"

Classifying contraceptives as preventive services and treating pregnant women as if they have fatal diseases is not a rational way to go about improving women's and maternal health outcomes.

4) The next all-natural blockbuster drug

Breastfeeding 90% of U.S. children for the first 6 months of life could potentially save the lives of more than 900 infants and $13 billion per year.

5) Breast cancer and the Angelina Jolie effect

Most women overestimate both their personal risk of developing breast cancer and the potential benefits of screening, driven in part by well-intentioned advocacy campaigns to raise "cancer awareness."

Abandoning risk factor assessment for HIV and HCV

During my family medicine training, I was taught about the importance of risk factor assessment in deciding what screening tests to order for patients seeking preventive health care. The principle is simple: to minimize false positives and their consequences, don't perform screening tests on patients who are unlikely to have the condition. For the chronic, treatable infections HIV and hepatitis C (HCV), adhering to this principle meant getting comfortable with asking patients about sensitive issues such as current or past illicit drug use and sexual behavior. For the latter, one of my teachers suggested a script along these lines: "Are you having sex? With one person or more than one? Male, female, or both?" Affirmative answers might lead to additional questions about specific sexual practices. Although patients occasionally give misleading or dishonest answers, self-reported drug use and sexual behaviors are considered accurate enough that these have long been accepted outcomes in research studies.

Risk factors are powerful predictors: for example, a person who has never used injection drugs or had sex with anyone who used injection drugs has a negligible risk of harboring asymptomatic HCV infection.

Given the above, I was surprised to read that the U.S. Preventive Services Task Force now recommends routine one-time screening for HCV in all persons born between 1945 and 1965, regardless of risk factors. In a significant change from the draft statement released last November, which suggested selective screening in this age group based on a somewhat higher baseline HCV prevalence related to the popularity of injection drug use in the 1970s and 1980s, the USPSTF decided instead to endorse the CDC's 2012 guideline and make screening the default choice for patients in the Baby Boom generation. This new recommendation comes on the heels of the Task Force's recent decision to abandon a risk factor-based approach to HIV screening and instead (like the CDC) test everyone between the age of 15 to 65.

Asked about the rationale for the USPSTF's change of heart, which caught the American Academy of Family Physicians off guard, Vice-Chair Dr. Michael LeFevre said: "We got a lot of public comments (on the draft recommendation), and many of them suggested that risk-based screening would essentially be a greater burden (to those doing the screening) than a birth-cohort approach, at least for that age group." He went on to cite 4 new studies that he said swayed the Task Force's change from a "C" to a "B" level recommendation, but since only one of those studies (a cost-effectiveness analysis, not a clinical trial) had anything to do with birth cohort screening, the new evidence was window dressing. The main issue is that talking with older patients about risk factors, which I consider to be an essential part of my training and practice, was thought by the USPSTF to be too burdensome a task for primary care physicians.

How have we really reached the point in this country where getting to know patients well enough to individualize their preventive health care is now considered a burden, rather than good medical practice? Primary care is considered so straightforward in many circles that independent nurse practitioners are thought to be as capable as physicians, but, apparently, it's also so complicated that we can't bother clinicians with taking a medical history. Sure, family doctors are sometimes harried or overwhelmed or forget to ask questions, but it's hardly a given that those same doctors will suddenly start ordering HIV and HCV tests for every Baby Boomer who walks in the door. More likely, prompted by preventive medicine process "quality" measures built in to electronic health records, many will simply add these on to the one-size-fits-all panels of mostly unnecessary tests that they already do at routine physicals, increasing health care costs for little discernible benefit.

A recent JAMA editorial by former USPSTF member Steven Woolf, MD, MPH and current USPSTF-AAFP liaison Doug Campos-Outcalt, MD, MPA asserted that the Affordable Care Act's mandate for health insurers to provide first-dollar coverage of the Task Force's "A" and "B" recommended services has had the unintended (and unwelcome) effect of politicizing the panel's decisions:

Placing the USPSTF in this position puts its analytic rigor at risk by preventing members from concentrating on the science. The inescapable reality that their conclusions dictate coverage guidelines may engender a subliminal pressure to reinterpret the evidence, knowing the ramifications, and to lower the threshold for A and B recommendations. Although it was founded to offer guidance to clinicians, the USPSTF must now rethink its guidance, knowing that every word in its recommendations constitutes a statutory mandate. ... In the struggle between scientific truth and political influence, the latter often wins.

Although the USPSTF's change of heart on HCV screening makes a lot of sense when viewed through this lens, there is no way for an outsider to establish with confidence that pressure from other federal health agencies, private advocacy groups, or companies that make lucrative drugs for HCV unduly influenced its final recommendation statement. Regardless of the Task Force's particular motivations, abandoning risk factor assessment for HCV and HIV in favor of universal age-based testing is not justified by existing evidence and, perhaps more importantly, undermines the time-tested approaches of family medicine and primary care.