A Qualitative Analysis of Lung Cancer Screening Practices by Primary Care Physicians

Susan Henderson; Amy DeGroff; Thomas B. Richards; Julia Kish-Doto; Cindy Soloe; Christina Heminger; Elizabeth Rohan

Disclosures

J Community Health. 2011;36(6):949-956. 

In This Article

Discussion

To our knowledge, this is the first qualitative study to formally explore physicians' perceptions of lung cancer screening. In our study, we identified seven factors influencing physicians' decision to screen for lung cancer: five physician-focused factors and two patient-focused. The physician-focused factors were (1) physicians' perception of the effectiveness of screening, (2) their knowledge of screening recommendations, (3) their practice experience, (4) their perception of a patient's risk for lung cancer, and (5) their fear of litigation. The two patient-focused factors were whether a patient requested screening and patients' ability to pay for the screening procedure (either out of pocket or via insurance). Because of these factors, we concluded that physicians often "struggle" in deciding whether to order screening tests for patients.

Our results illustrate the complexity involved in physicians' decisions whether to screen patients for lung cancer. Although most physicians participating in our focus groups seemed to be familiar with the lung cancer screening guidelines of major North American organizations, they often ordered screening tests on the basis of other factors. Results of previous studies have similarly shown that many physicians do not follow cancer screening guidelines. For example, in a national study involving 6,712 patients, researchers found that of 41 indicators for screening (for breast cancer, cervical cancer, smoking) guidelines were followed only 52.5% of the time, even though they were standards of care.[17] Changes in screening guidelines, as occurred recently in screening guidelines for breast cancer,[18] may be another factor limiting the extent to which physicians adhere to them.

Some of the physicians in our focus groups reported screening what they called "high-risk" patients, although their definitions of "high-risk" varied substantially. Criteria they reported using to determine whether patients were at high risk for lung cancer included the number of packs of cigarettes smoked, environmental exposures, and family history. Results of other studies have similarly shown that physicians often use imprecisely defined determinations of risk to determine whether patients should be screened for cancer, including one study in which more than half of US pulmonologists and thoracic surgeons were found to believe that "high-risk asymptomatic patients" (not defined) should be screened for lung cancer.[19] In another study, Burger et al. found that 29% of US radiologists believed that all patients with at least a 20-pack-year history of smoking should be screened for lung cancer with CT scans, and that 42% believed that patients should be screened "if [screening] was very important to [them]".[20] Our data provide some insight as to why physicians may screen high-risk patients regardless of guidelines. In particular, a recurring theme is that current medical practice is more patient-centered, that is, the responses of physicians in our focus groups indicated that some may accommodate patients' requests for lung cancer screening tests if they perceive the requests to be reasonable, whether or not they are indicated tests. In addition, this may be influenced by defensive medicine, in which physicians will order tests to protect themselves from possible future litigation. Our finding that several physicians reported agreeing to order a test when a patient asked for it even if it was not indicated suggests that physicians are either too busy to communicate appropriately with their patients or have inadequate training in shared decision-making, in which a patient and a physician jointly decide a course of action. The ACS has recommended informed decision making as part of its effort to make screening more patient-centered. We did not explore this important distinction, in which the individual is consulted and a joint decision is made, as opposed to being physician-requested alone. It was concerning that some physicians did not differentiate between screening and diagnosis. This suggests that physicians may need more public health education, both in medical school and as part of their continuing medical education. Physicians should understand that the goal of screening is not simply to detect more cancers, but to detect cancers at an earlier stage when they are more treatable.

Limitations

Because only 28 physicians participated in our focus groups and participation was limited to physicians who were accessible to the recruiting center, the results are not generalizable to all US primary care physicians. In addition we did not ask physicians whether they had X-ray facilities in their clinics and thus stood to profit from screening, did not include nurse practitioners and physician assistants in the focus groups even though they are also authorized to order lung cancer screening, did not include physicians employed at academic centers or government-run facilities, did not assess the possible relationship between physicians' screening practices and where they trained or the number of years that they had been in practice, and did not assess the possible effects of social desirability biases on physicians' responses.

Recommendations

On the basis of our findings, we offer four recommendations.

  1. Ensure that clinicians are educated about lung cancer screening guidelines and the potential advantages and disadvantages of screening beginning in medical school, and that they are trained to counsel patients on tobacco use prevention and cessation.

  2. Educate patients about the purposes of lung cancer screening and when it is appropriate and encourage them to participate in shared decision-making about screening with their physician.

  3. Conduct further exploratory research about how physicians handle patients' requests for lung cancer screening tests.

  4. Conduct larger, quantitative studies to better understand how physicians use lung cancer screening tests and how patients perceive lung cancer screening. Such studies should attempt to determine how various factors such as physicians' years of experience and their fear of litigation may affect their decisions to screen patients for lung cancer.

Screening is a complex process involving both the patient and the patient's physician. As previously noted,[21] patients and the general public should be made aware of controversies about cancer screening so that they will be more willing to work with the medical community to resolve them. Although results of recent research suggest that use of CT scans to identify people with lung cancer may contribute to a reduction in lung cancer mortality rates,[12] other factors, such as those we identified, may affect the feasibility of using CT scans to screen for lung cancer. For now, the one proven means of reducing the incidence of lung cancer is the implementation of comprehensive tobacco control programs.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....