Lung Cancer Screening Practices of Primary Care Physicians: Results From a National Survey

The article by Klabunde et al. is quite sobering. Clearly, screening for a deadly disease such as lung cancer is needed, but at this time, despite findings from the NLST, the benefits of low-dose spiral computerized tomography (LDCT) may not be sufficient to warrant the possible harms. Most organizations do not endorse any type of screening for lung cancer in average or high risk individuals, be it with a CT scan, chest x-ray, or sputum cytology. Yet, this study revealed the apparent willingness (based on clinical vignettes) of primary care providers to advise patients to undergo a variety of lung cancer screening modalities. Further, the primary care providers also acknowledged using these modalities in clinical practice.

The findings of this study not only suggest that primary care practice is not evidence-based with respect to lung cancer screening, but also point to missed opportunities in allocating time and resources in a busy clinical practice. Most cases of lung cancer are preventable. Prior literature has shown that primary care physicians attribute inadequate tobacco cessation counseling to "lack of time;" yet, time does not appear to be a barrier for many of the respondents to recommend that patients undergo a chest x-ray and/or LDCT, the latter associated with a high rate of false positive findings. As a primary care provider, I spend quite a bit of time explaining incidental findings, ordering and reviewing follow up imaging, and then explaining some more. This time could be better invested in counseling patients to quit smoking and thereby prevent not only lung cancer but also a host of other cancers and deadly diseases.

As for evidence based practice... evidence evolves over time; the recent changes in cervical and prostate cancer screening recommendations emphasize that point. Other organizations may soon follow NCCN guidelines and recommend LDCT screening among high risk patients. Yet primary care physicians should remain cautious and not be overly zealous to embrace any screening modality (whether for lung or other cancers) in attempts to "catch disease early." It may not be in the patients' best interest.

Competing interests:   None declared

It is always an eye-opener to learn what real-world doctors believe and say they do. Too often we learn about substantial overuse, underuse, or misuse relative to the best available research evidence of effectiveness. In the Klabunde paper,(1) 57% of the primary care physicians had ordered some form of lung cancer screening at a time when there was evidence that screening did not prevent lung cancer deaths (chest x-rays and sputum) or no strong evidence that it did (low-dose CT scanning) . Overuse is a problem, whether or not screening prevents lung cancer deaths, because this screening does produce many false positive results that lead to extra procedures with attendant risk, hassles, worry, and costs.

Now, a few years after the survey, there is randomized controlled trial evidence that low-dose spiral CT does reduces lung cancer mortality.(2) Even so, there is so far no consensus on lung cancer screening; expert groups are wrestling with whether effects are large enough to justify a substantial number of false positive results and extra testing, as well as financial costs, and whether they should generalize results from patients in the trial, who were at increased risk, to a larger proportion of adults at somewhat lower risk.

As for risk factors for overuse, some of the findings seem circular: physicians are more likely to recommend screening when they believe it works and that guidelines recommend it. Others are more informative. Older doctors are at greater risk of getting it wrong, as has been shown in a variety of studies. (3) Internists provide more aggressive care than family physicians, which is also consistent with other evidence.

What is the way out of this situation, and others like it, where quality and cost are so poorly related to evidence? Scholars think first of education but evidence suggests that it is not a particularly strong way of changing physicians' behaviors. Health systems need to invest more resources into other efforts to change behaviors such as financial incentives/disincentives, use of local opinion-leaders, and reminders. Malpractice reform could buffer physicians from fears of well-considered inaction even in the face of patients' requests, when the evidence favors that approach. Cost associated with everyday overuse is so high that such an investment might even result in net savings.

References
1. Klabunde CN, Marcus PM, Han PK, et al. Lung cancer screeing practices of primary care physicians: Results from a National Survey. Ann Fam Med 2012;10(2):102-110.
2. The National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365(5):395-409.
3. Chaudhry NK, Fletcher RH, Soumerai SB. Systematic review: The relationship between clinical experience and quality of health care. Ann Intern Med 2005;142:260-273.

Competing interests:   None

LUNG CANCER SCREENING --- still a foggy forecast

The question of whether - or how--to screen for lung cancer is among the most vexing in modern pulmonary practice. The topic is especially important since lung cancer is by far the most common lethal cancer in the United States; and it is tempting to postulate that screening might be successful since we can easily identify the at-risk population (smokers) and readily detect small tumors non-invasively with increasingly sophisticated imaging systems. A series of multicenter studies in the 1980's showed that both chest x- rays and sputum cytology were ineffective for lung cancer screening, though some of the methods and conclusions were bitterly contested. Thus the publication of the National Lung Screening Trial (NSLT) in 2011, showing a 20% reduction in lung cancer mortality with low-dose CT screening, was greeted with real excitement and enthusiasm. Interestingly, the results suggested that each of three yearly screens disclosed new tumors; thus a net benefit might attend long term screening, at least until deaths from coexisting chronic diseases limit the gains in life expectancy. However, the authors cautioned that current NLST data are insufficient to establish lung cancer screening recommendations and noted that "before public policy recommendations are crafted, the cost effectiveness of low-dose CT screening must be rigorously analyzed." And given our current national politically charged debate about health care costs, paying for a large-scale national lung cancer screening program will likely become even more contentious. Against this backdrop the study by Klabunde and coauthors in the current issue of Annals of Family Medicine is of particular interest. The authors present the results of a national survey of 962 primary care physicians with regard to lung cancer screening practices. It is the first such survey in decades, and provides a useful snapshot of practice patterns in 2006-2007, long after the chest X-ray/sputum cytology studies of the 1980's but well before the NSLT results were published. Among other findings, Klabunde et al report that 25-50% of physicians ascribe to misconceptions including that expert groups recommend lung cancer screening, chest x-rays effectively lower lung cancer mortality, and sputum cytology lowers lung cancer mortality in former or current smokers. In summary, we have a vital medical issue characterized by substantial physician misinformation and improper test utilization, while the relevant database continues to evolve and authoritative guidelines are in flux-- truly a recipe for bad medicine. In my opinion only a multifaceted approach can extricate us. This must include continuing research to refine the use of screening tests and formulate easily applied screening strategies, massive ongoing physician education, and probably incentive programs to reward optimal test utilization. Bring on the sun!

Competing interests:   None declared

I am responding to the note submitted by Dr. Hadyn Williams regarding the article, Lung Cancer Screening Practices of Primary Care Physicians: Results from a National Survey, published in the March/April issue of Annals of Family Medicine. Dr. Williams is correct that the National Comprehensive Cancer Network produced new lung cancer screening guidelines in 2011. However, these guidelines came out after publication, and as a result of, the National Lung Screening Trial results. The survey that Annals of Family Medicine study that I led was based on was conducted in 2006-2007, and queried primary care physicians about the lung cancer screening guidelines that were in place at that time (which the NCCN ones were not). It should be emphasized that the NCCN guidelines Dr. Williams cites are brand-new, and were not assessed in our 2006-2007 survey, nor is there any published evidence at this time regarding adoption of these new guidelines. Prior research has shown that primary care physicians pay particular attention to cancer screening guidelines from the U.S. Preventive Services Task Force and American Cancer Society; neither group has yet updated its lung cancer screening recommendations.

Competing interests:   None declared