The CMS Requirement for Shared Decision Making for Lung Cancer Screening

In 2011, the National Lung Screening Trial (NLST) demonstrated that annual screening with low-dose CT (LDCT) reduced lung cancer mortality among current or former heavy tobacco users.¹ While the trial found that LDCT screening was beneficial, screening also posed important harms, including false positive results, radiation exposure, overdiagnosis, incidental findings, and complications from invasive procedures, particularly in participants who did not have lung cancer. The potential complications from the invasive procedures, which included percutaneous needle biopsy, bronchoscopy, and thoracic surgical procedures, are far riskier than those associated with other cancer screening programs.

In 2015, a Decision Memo from the Centers for Medicare & Medicaid Services (CMS) determined that LDCT was a covered benefit for eligible adults, but stipulated that beneficiaries first undergo a billable counseling visit for shared decision making (SDM) using a patient decision aid.² This unprecedented mandate for ordering a screening test recognized that individuals differ in balancing the importance of gaining benefits versus avoiding harms. The counseling visit should enable the values and preferences of the informed patient to be elicited and systematically incorporated into the screening decision-making process. With the USPSTF now recommending expanded screening for lung cancer to include populations at lower absolute risk of lung cancer, the importance of SDM has increased. (JAMA, ref, to be added).

Lung cancer screening uptake has increased in recent years, albeit slowly. Many experts find screening rates to be unacceptably low, and some have argued that the expectation for a SDM visit is a barrier to screening.³ Clinicians do cite limited time for discussions, competing clinical demands, lack of decision aids, and inadequate system and staff support for implementing SDM in routine clinical care--and thus to initiating lung cancer screening.⁴ However, attributing low screening rates to the required SDM visit is a misplaced criticism because it ignores the fact that non-trivial numbers of screened patients will experience harms, including the prospect of high out-of-pocket costs. The principle of respect for patient autonomy obliges the medical community to inform patients about the tradeoffs involved in the decision to be screened for lung cancer. The purpose of SDM visit is not to impact overall screening rates, but rather to ensure that the individual values and preferences of an informed patient are part of the screening decision.

CMS should continue requiring and reimbursing counseling visits for the consequential decision regarding initiating LCS. The importance of these discussions is supported by increasing evidence that informed patients have different preferences about screening, particularly after participating in SDM in a primary care setting.⁵ Shared decision making with a patient decision aid has been shown to increase knowledge of the tradeoffs between benefits and harms of screening decisions, improve perceptions of risk, improve communication with clinicians, reduce decisional conflict, and help clarify personal values.⁶ A recent systematic review of SDM interventions for lung cancer screening found that they increased patients’ knowledge, reduced decisional conflict, and were highly acceptable to patients and providers.⁵ The counseling visits are also expected to address tobacco use and guide patients towards the tobacco abstinence and cessation interventions strongly recommended by the US Preventive Services Task Force (USPSTF) for primary prevention of lung cancer and other tobacco-related illnesses. By reimbursing this visit, CMS helped address an important barrier to supporting patients in making complex decisions–the lack of financial incentives for SDM.

Shared decision making may improve health equity around lung cancer. The new USPSTF recommendation increases the number of eligible patients, including women and underserved minorities, particularly Black individuals who have the highest rates of lung cancer incidence and mortality. While broadening the eligibility criteria may help reduce disparities in lung cancer mortality, engaging vulnerable populations potentially may be difficult due to distrust of the medical system, feeling stigmatized about smoking, having lower levels of health literacy, and being at risk for financial toxicity from medical care.⁷ Providing high-quality decision support is a patient-centered approach that may potentially increase either screening uptake or tobacco cessation among vulnerable populations. Both of these outcomes can lead to reducing lung cancer mortality. Using a patient decision aid as currently required by CMS may help ensure high-quality SDM by consistently providing patients comprehensive, balanced, and clear information about the benefits and harms of screening. However, these tools need to be tailored for culturally diverse populations and those with low health literacy and numeracy.

Another benefit of SDM is the potential to improve adherence among those choosing to enter a screening program. Without patient adherence to annual testing and to recommended diagnostic procedures and treatment, screening programs will not be able to reduce lung cancer mortality and may result in net harms. The participants in the NLST were 95% adherent with the three rounds of annual screening and 90% of those with a localized cancer underwent curative surgery. Expecting similar adherence in non-research settings is not realistic. A recent meta-analysis of studies conducted in either academic or community settings estimated that LCS adherence following a baseline screening was only 55%.⁸ Although population-based data supporting the role of SDM in longitudinal adherence to lung cancer screening are lacking, there is some evidence from randomized trials that patients exposed to decision aids have greater adherence with their decisions.⁶ Given that substantial proportions of patients decline LCS after SDM, the requirement for SDM may select for motivated patients who are more likely to be adherent with screening programs. This hypothesis, though, needs to be rigorously evaluated in clinical trials.

The optimal delivery structure and personnel for providing SDM in practice remains uncertain and is a substantial challenge. Health systems such as the Massachusetts General Hospital have successfully introduced the systematic use of decision aids for preference sensitive conditions into patient care.⁹ However, payor restrictions on who can deliver SDM inhibit broad implementation. CMS expects that the SDM visit be conducted by a licensed independent practitioner (LIP), which includes physicians, physician assistants, or nurse practitioners. Ideally, SDM should be accomplished in primary care because primary care clinicians are best positioned to determine whether a patient is healthy enough to undergo LCS and to prioritize screening in the setting of other competing health issues. However, the expectation that the primary care clinician deliver SDM is widely recognized to be unrealistic given time constraints, lack of resources to support integrated decision processes, and limited familiarity with guidelines and screening trial results. Evidence suggests that clinician discussions around screening can be quite abbreviated, often emphasizing benefits over harms and failing to elicit patient preferences.^10,11 Training a vast number of clinicians to be able to deliver a limited SDM discussion during an already over-busy clinic visit does not seem to be a feasible approach for delivering high quality LCS decision support. A more practical strategy would be to use trained non-physicians who have a dedicated role for LCS and sufficient time to conduct an SDM visit. This would help ensure patient engagement, deliver consistent and comprehensive information, and support tobacco cessation. Non-physician decision coaches are acceptable to patients and can increase values-concordant decisions and satisfaction with the decision process.¹² CMS should reimburse the efforts of non-LIP because health decision coaches, who can include nurses, psychologists, and social workers, can effectively deliver SDM. Centralizing the process may also enable health care systems to proactively identify and engage a more inclusive population of eligible patients outside of scheduled clinic visits.

The context of SDM delivery is also important. The COVID-19 epidemic has substantially increased the use of telehealth services. Telehealth visits offer important advantages for SDM for lung cancer screening and CMS should continue this coverage. In addition to convenience, telehealth delivery gives patients relevant information early in the decision process and facilitates patients in making an LCS decision in a primary care context without requiring a face-to-face visit. Evidence suggests that populations of screen-eligible patients who receive decision support in primary care settings have varying screening preferences, whereas nearly all patients who attend a screening clinic for SDM (when a CT has often already been scheduled) are past the “decision window” and proceed to screening.⁵

To summarize, SDM for lung cancer screening is a valuable, ethical, and effective strategy for supporting a patient-centered approach to this highly consequential decision. SDM should remain an expected component of the screening process. The time and resource-constrained clinician visit remains a major impediment to implementation, and both the quality and feasibility of SDM visits could be enhanced if CMS and other payors would cover SDM visits delivered by an expanded pool of trained, non-physician staff, particularly through telehealth. Screening can reduce lung cancer mortality only if patients are committed to the screening process, including performing initial and annual screening tests, controlling tobacco use, and undergoing recommended diagnostic testing and treatment.