Should a 60-year-old who's smoked a pack of cigarettes a day for 30 years be screened for lung cancer using low-dose computed tomographic (CT) scanning? What about a 42-year-old woman with no family history of breast cancer: should she undergo mammography? Should prostate-specific antigen testing be performed in any but the highest-risk patients, given the high rate of false-positive results and the harms of subsequent tests and treatments?
All of these screening procedures, and others, have provoked controversy in recent years. In this era of evidence-based care, many clinicians and patients expect clear guidelines outlining what represents “best practices,” but a sharp divide exists when it comes to screening asymptomatic patients. Which screenings should be done, and in whom?
Lung cancer provides a stark example; it's by far the leading cause of cancer death in the United States, and some have hoped that screening with low-dose CT scanning might detect the disease at earlier, more treatable stages. Recently, supporters and opponents of this screening method have voiced their positions in an unusually public way.
- In 2012, the American Lung Association recommended screening in long-term smokers, citing the National Lung Screening Trial (NLST), results of which suggested that the $300-to-$500 procedure “could reduce” lung cancer deaths by 20%, compared with X-ray (see http://bit.ly/1tu9xex).
- Last December, the U.S. Preventive Services Task Force (USPSTF) cited the NLST in granting a “B” recommendation for the screening to be used in asymptomatic high-risk older smokers, indicating a “moderate certainty” that its benefits outweigh its harms (for the USPSTF analysis, go to http://bit.ly/1cUEu8z).
- In January, the American Academy of Family Physicians opposed the USPSTF (http://bit.ly/JYyZJI), issuing an “I” recommendation for the scans, concluding that there's insufficient evidence to support their use in lung cancer screening and expressing “significant concern with basing such a far reaching and costly recommendation on a single study.”
- In February, it was reported in JAMA Internal Medicine that 18% of cancers detected by CT scanning in the NLST were actually minimally invasive adenocarcinomas and that such “potentially clinically insignificant” cases likely resulted in “overdiagnosis”—in other words, inaccurate diagnoses of lung cancer and unnecessary invasive procedures.
In April, the Medicare Evidence Development and Coverage Advisory Committee advised against Medicare coverage of this screening, saying the benefits don't outweigh the harms (a final decision will be made next year). A bipartisan group of 42 U.S. senators has petitioned Centers for Medicare and Medicaid Services administrator Marilyn Tavenner to make the determination in favor of Medicare coverage for the screening—and to make it “expeditiously.” Complicating the argument are reports of a study suggesting that the procedure could cost billions of dollars per year and increase Medicare premiums by $3 to $9 per month (http://usat.ly/TcYknG).
A close look at most recommendations on health screenings, including those supporting controversial procedures, shows that they encourage clinicians to take a shared–decision-making approach. But what does that mean in actual practice for nurses and patients?
Overreliance on Screening vs. Shared Decisions
Marie Truglio-Londrigan has been studying shared decision making—whether it works for patients and what prevents nurses from engaging patients as partners in their care. A professor at the College of Health Professions, Lienhard School of Nursing, in Pleasantville, New York, Truglio-Londrigan said she's discovered in her research that too often nurses assume that patients have preferences for screening, when in fact many need more information on all options available to them. “What's the test looking for? What are the pros and cons? Then they can see how the end decision fits into their lives.”
Shared decision making represents a democratization of health care, in Truglio-Londrigan's view, but she said that a culture of maternalism and paternalism in nursing works against that goal. “Nurses speak a good line about shared decision making, but often they think they know what's best for the patient and they want to protect the patient.”
Truglio-Londrigan recognizes that many health care organizations don't support shared decision making in nursing care. Still, she believes that nurses should try to “listen deeply,” especially when a patient seems to resist talking about a particular screening procedure.
Randy Jones, associate professor and assistant director in the Office for Nursing Research at the University of Virginia, acknowledges that the health care system could do more to engage patients more fully in screening decisions. Jones has studied African American men who're facing decisions around prostate cancer screening and treatment; African Americans are disproportionately affected by many conditions, the best publicized being heart disease, and they die at higher rates than whites from many cancers.
Jones and his colleagues are studying the effects of community and nurse “navigators” on decision making among patients living in rural areas. The goal is to provide “different layers of contact” within the health care system. The community navigators are trained local laypeople who assist patients with financial issues, making and keeping appointments, and accessing needed resources. The nurse navigator takes over from the community navigator for clinical questions and decisions, but the goal is simple: to enhance trust, and thereby augment decision making, among rural and minority patients.
“In the minority population, trust is a big issue [when it comes to] cancer and screening behaviors,” Jones said. “The patient forms a relationship with the navigator and feels like, ‘Yes, the system is looking out for me.’”
To Gene Harkless, it's clear that health screenings “are not the salvation” many wish them to be. She's chairperson of the Department of Nursing at the University of New Hampshire in Durham and an NP at Families First, a federally qualified health center in nearby Portsmouth, and she believes that changing the health care system's overreliance on screening and other unnecessary procedures will require a “resocializing” of clinicians, including nurses.
“Nurses learn how to run a new piece of equipment or how to care for that wound, but very little attention is paid to communication,” Harkless said recently. “We don't focus enough on the basics—how to listen, how to give feedback.” Even the teach-back method—a standard of nursing education for decades that involves asking patients to repeat in their own words what they understand about their condition and options for screening or treatment—has been squeezed out of the 15-minute primary care appointment, she said. But as a teacher of nursing students, she sees reason for hope.
“Students are learning to have those conversations. We're moving toward a more critical group of nurses. The next generation is going to look very different.”—Joy Jacobson
Choosing Wisely, an American Board of Internal Medicine initiative, aims to encourage dialogue between clinicians and patients in choosing care that's evidence based: www.choosingwisely.org.
Option Grids are “brief, easy-to-read tools” that can help consumers and clinicians compare treatment or screening choices: www.optiongrid.org.
“Too Much Medicine,” a special issue of BMJ, will be published to coincide with the second Preventing Overdiagnosis conference, which will take place September 15–17, sponsored by the Centre for Evidence-Based Medicine at the University of Oxford, England: www.bmj.com.
Decision aids have been shown to be effective in raising awareness of the pros and cons of a variety of procedures. See a systematic review in which Stacey and colleagues examined 115 studies on the effects of decision aids on screening and treatment decisions: http://bit.ly/POudky.
The Patient Decision Aids Research Group presents the research evidence related to many decision aids, including several for screenings: http://bit.ly/1fTUdoo.