Questions and Answers: AHRQ National Webinar on Shared Decisionmaking Tools for Lung Cancer Screening
Wednesday, May 4, 2016
On May 4, 2016, the Agency for Healthcare Research and Quality (AHRQ) hosted a national webinar on the topic of Shared Decisionmaking Tools for Lung Cancer Screening. The presenters and moderator of this webinar are identified below. This document includes the questions posed by webinar participants and the responses to those questions by the presenters and AHRQ.
- Robert Volk, Ph.D., University of Texas MD Anderson Cancer Center: Evidence-based Policy Recommendations about Shared Decisionmaking for Lung Cancer Screening and Introduction of an Implementation Toolkit that Meets Current Policy Requirements
- Richard Street, Jr., Ph.D., Texas A&M University and Baylor College of Medicine: Practical Strategies for Implementing High Quality Shared Decisionmaking Processes in the Primary Care Setting
- Monique D. Cohen, Ph.D., M.P.H., Agency for Healthcare Research and Quality (AHRQ)
Low Dose Computed Tomography (LDCT) Shared Decisionmaking Implementation
Communication Strategy Questions
Age Range for LDCT Screening-Related Questions
CMS Policy-Related Questions
Question: Is there a cost to access the toolkit?
Monique Cohen: No. The toolkit is available for free. You can access it online through the Effective Healthcare Web site at: https://www.effectivehealthcare.ahrq.gov/tools-and-resources/patient-decision-aids/lung-cancer-screening/patient/
You can also order print versions through the AHRQ Publication Clearinghouse. The number is 800-358-9295. I'm not sure if the printing has been completed yet. But you can place your order now, and they will send you the toolkit as soon as it's ready.
Question: Are the toolkit and the patient-focused products available in Spanish?
Cohen: They will be available in Spanish. They're currently being translated.
Question: Regarding the development of the decision aid, can you talk a little bit about how long it took to develop the resources and what types of clinicians provided input?
Robert Volk: The toolkit was developed during one contract year. The development process spanned about nine months of that. There was some upfront work that had to be done as we selected topics, but we did all this within a single contract year.
Before we launched development, we talked with some practicing primary care clinicians about their concerns with implementing the new recommendations, their challenges, and what they thought would be particularly helpful in creating an implementation toolkit.
We also surveyed a large number of primary care clinicians about what they are doing now. That study occurred after the Task Force its issued updated recommendations and at about the same time that the Centers for Medicare and Medicaid Services (CMS) was completing its final national coverage determination. The draft policy was already out, and people could have been aware of that. We obtained surveys from about 300 primary care clinicians. We found that a small minority already had lung cancer screening programs in place, but they had some significant needs around implementation. They were practical things, like the need for patient decision aids and patient education materials, and about how to find a certified radiology center.
Question: What mechanisms or plans are in place to make sure that this toolkit stays updated?
Volk: It's important to have an updating policy in place when someone releases a patient decision aid. Our updating policy for this decision aid is available online at: https://effectivehealthcare.ahrq.gov/ehc/assets/File/lung-cancer-about-160226.pdf
We might update the tool if new evidence comes out about the harms and benefits of lung cancer screening. This might trigger us to look again at the evidence base that informs the task force recommendations and any of the products that we have developed as a result of them. Also, it's good practice to look at the literature and the tool every one to two years to make sure the messaging is still on target and that the tool reflects the literature appropriately.
Question: What was the role of patients in the development of the toolkit?
Volk: We sought patients' input as we considered the messaging that we wanted in the tool. For example, that's where the issue of providing information about insurance coverage came up. Patients were adamant that they wanted to know about who was going to pay for what.
Once we had a prototype available, we did focus cognitive testing with patients where they would look at a mock-up of the tool and tell us, in their own words, what the information was trying to convey. If the messaging wasn't quite right, we did some tweaking. Some concepts proved challenging for us to explain. For instance, some patients struggled with the concept of over-diagnosis. They also confused it with the concept of a false positive. We're still trying to figure out how to convey those concepts.
We also did acceptability testing. We asked patients if they thought the tool was helpful, clear, and the right length. We also asked them if they thought it presented information in a balanced fashion. We made changes iteratively throughout that process.
Question: Based on the work and testing that you did with patients, would you say that the tool is appropriate for use with patients who have lower health literacy? Are there any strategies that you would suggest for clinicians who have a lot of patients with low health literacy?
Volk: The patient decision aid has a lot of information in it. For example, the icon array is very difficult for someone to negotiate on their own if they're not used to thinking about numbers and comparing rates, such as for screening and not screening. So it's important for clinicians to use some of the strategies in the checklist—especially teach-back—to probe the patient's understanding and clarify some misconceptions. The Encounter Tool can also be helpful with that, because the clinician and patient can look at it together.
Question: In developing the toolkit, did you receive any input from staff at CMS?
Volk: Yes. We sent the tool through CMS and got some helpful feedback. They felt that the intervention was consistent with beneficiary eligibility requirements.
Question: Do patients tend to prefer having shared decisionmaking conversations with one type of clinician versus another?
Volk: I'm not sure if any research addresses a preference for the type of clinician.
Question: How many patients might decline screening following shared decisionmaking? Or is it still too early to tell?
Volk: It's really too early to tell. When we look back at some of the other cancer screening questions, it's kind of a mixed bag. For example, when patients participate in interventions that provide information about the harms and benefits of prostate cancer screening with the prostate specific antigen assay, there is a reduction in their interest in PSA testing and in their actual completion of testing. But the use of patient decision aids seems to increase the colorectal screening rate. If I had to venture a hypothesis about this, smokers know that cancer is a very real risk for them, and they're concerned about it. So I'm guessing that the rate at which people decline will be fairly low. But again, it is too early to really know.
Question: How long would you expect a counseling visit to take for a clinician who uses the resources from the toolkit?
Volk: I don't really know how long the visit would take. We haven't looked at that yet. I think it's going to vary, in part based on the patient. If somebody comes in with a lot of concerns or has little information about lung cancer screening, the conversation may take more time. However, the patient decision aid could help the patient prepare for the conversation.
I think the concern driving this question is that primary care clinicians really don't have a lot of time to have these conversations with patients. Some of this work could actually be done by other clinicians, such as nurses. That would take some of the pressure off physicians.
Question: Do you have any suggestions for getting patients to come in for an initial visit? Also, who would do the prep work? Would it be an RN or a medical assistant? Do you use a verbal or written invitation for the appointment?
Volk: Some clinicians send letters to patients who smoke, to invite them to come in for screening. Others talk to patients about screening while discussing smoking cessation during a periodic health exam. Clinicians could also tell patients about screening during a group shared decisionmaking visit.
Question: Do you have any tips for how to communicate the high positive or false alarm rate to patients?
Volk: People need to have a lay understanding of what predictive value is all about. Most abnormal lung cancer screening scans are not lung cancer. Only about 1 in 20 actually turn out to be lung cancer.
Also, the vast majority of people who have a positive result either get additional imaging, or are simply watched. Only a very small number go on to have invasive procedures. Those numbers are shown in the pictographs. In talking about the high false-positive rate, it's important to emphasize that these are the consequences that result from it.
Question: How do you respond to concerns that people might interpret the shared decisionmaking visit as trying to talk eligible patients out of being screened?
Volk: In shared decisionmaking, the patient makes an informed decision about screening based on the best available evidence during a conversation with a healthcare provider that includes consideration of what is most important to the patient. A patient making an informed decision after a conversation with their provider is the desired goal.
Richard Street: Getting the patient to talk about what they know/think first is important, because it tells you their starting point. It's also important to have a strong relationship with your patient.
Question: When a patient asks me what their chances are of being cured from lung cancer, what should I say?
Volk: The lung cancer-specific mortality reduction from the NLST was about 20 percent. Mortality after 6½ years drops from 21 in 1,000 individuals not screened to 18 in 1,000 individuals screened with LDCT. It is still the case that most people diagnosed with lung cancer will die from it.
Question: Since both groups in NLST received screening, there are no statistics available regarding benefits of lung cancer screening vs. no screening. The icon array is somewhat misleading using the comparison titled “not screened with LDCT.”
Volk: We had to make some simplifying assumptions when creating the icon arrays. It is true that the NLST comparator was a standard chest X-ray. But a chest X-ray is not an effective strategy for lung cancer screening, so we decided to present it as not being screened. Others have done the same. There is always a tradeoff between specificity and presenting so much information that patients become confused.
Question Could you talk a little about the differences in the upper age limits between the task force and CMS recommendations?
Volk: The U.S. Preventive Services Task Force (USPSTF) used age 80 as its upper limit. That number was based on some of the comparative modeling they did that showed there was a benefit in screening patients up to age 80.
The National Lung Screening Trial (NLST), however, included only patients up to age 74. The USPSTF does have higher age limit than what was used in the NSLT.
The CMS beneficiary eligibility includes screening up to age 77. My understanding is that CMS wanted people who started screening at the age of 74 to get the full 3 years that people in the NLST received. I think that's the reason for the upper age limit.
Question: What should be done in treating those patients who fall below the age 55 minimum or above the age 77 maximum?
Volk: According to the CMS Policy Statement, they're not eligible for lung cancer screening. With those patients, it's important to enforce the messaging around smoking cessation and smoking abstinence. Those are the most important things you can do to lower your risk of developing and dying from lung cancer.
Question: What do you think about doing these kinds of visits using TeleHealth?
Street: I would suggest having a live video conference. You can do this to share or explore information, or to discuss patients' concerns and options. The key is to make sure both you and the patient have a copy of the aid to be used in the encounter. Ideally, the patient has already filled it out, so you can go over it together.
Question: What do you think about having group shared decisionmaking visits (in which there is a group discussion/presentation and then a chance to ask individual questions) about lung cancer screening?
Volk: I like the idea. I believe it may offer some efficiencies. But it is my understanding that CMS will not cover a group visit model.
Question: What is a reasonable decline rate after the shared decisionmaking conversation?
Volk: The experience from the Cleveland Clinic lung cancer screening program is less than 5 percent of patients who participate in a shared decisionmaking visit decline screening. Patient choice appears to be a major driver. But if choice reflects an unwillingness to have surgery, this has more to do with screening eligibility than participating in a conversation about harms and benefits. The decline rate may be even lower.
Question: Why is reduction in disease-specific morality from this screening 3/1K, while reduction in overall morality is 5/1K?
Volk: A careful reading of the NLST results shows that the overall mortality reduction is driven by lung cancer-specific mortality reduction. When the lung cancer deaths are pulled out, mortality for the chest X-ray and LDCT groups is not significantly different. Some people argue that LDCT has the potential to identify other, non-cancer disease, and this may explain the reduction in overall mortality, but currently there is no data supporting that claim.
Question: How have patients reacted to the tools? Does there seem to be a preference for one versus the other? How can and/or should the tools be used together?
Volk: We don't have information about a preference for one or the other. They certainly can be used together. The Encounter Tool is particularity attractive, because we do want to help foster a conversation between the healthcare provider and the patient. The patient decision aid can help a patient prepare for a visit. Using the two together could be helpful.
Street: We don't have evidence that one approach is better than another. These studies haven't been done yet. Both have value and can be complementary. The Encounter Tool affords the opportunity to explore any misconceptions that the patient might have and talk about what's important to the patient.
Question: What kinds of clinicians, other than physicians, can engage patients in a shared decisionmaking visit and still be covered?
Volk: In its updated billing code, CMS talks about the visit being furnished by a physician or a qualified physician—meaning a physician assistant, a nurse practitioner, or a clinical nurse specialist. They have a specific citation for that. Again, I direct your attention to the link to the CMS policy statement in the slides (https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=274). You may want to take a look at that to get further clarification on those issues.
Question: Is an education visit (separate from the screening visit) required to order the LDCT?
Volk: No. That is the same visit. The order can be placed at the time the patient participates in the patient counseling and shared decisionmaking visit.
Question: Does the shared decisionmaking need to take place after the first year of screening? Or is it required only once?
Volk: If we're talking about the CMS eligibility criteria, shared decisionmaking is required for the first referral.
Question: Can this decisionmaking discussion take place during a routine office visit? Or does it need to be scheduled as its own appointment?
Volk: We need to get some guidance from CMS about that.
Question: Are radiologists included in the list of individuals who can provide the shared decisionmaking visit?
Volk: I do not know the answer to that. That question is best directed to CMS.
Question: Does the shared decisionmaking documentation need to come from the referring provider, even though other clinicians may be involved in the process?
Volk: That's my understanding. The documentation does come from the referring clinician.
NOTE FROM AHRQ: There were a number of CMS policy-related questions that neither the presenters nor AHRQ were able to address related to:
- Counseling and shared decisionmaking visits.
- Qualified providers.
- Smoking cessation counseling.
- Auditing/quality of LDCT facilities.
Below, we provide published answers from CMS to some of the common questions arising during the webinar. For more information, please see:
- Decision Memo for Screening for Lung Cancer with Low Dose Computed Tomography (LDCT) (CAG-00439N): https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=274
- CMS Medicare National Coverage Determinations Manual. Chapter 1, Part 4 (Sections 200 – 310.1), Coverage Determinations (revised 2/5/2016): https://www.cms.gov/Regulations-and-Guidance/Guidance/Manuals/downloads/ncd103c1_part4.pdf
CMS Language About Counseling and Shared Decisionmaking Visits
It appears that the initial visit must be a specific event, but that subsequent screenings can be incorporated into other visits. CMS Decision Memo, page 37 of the PDF, states:
“We agree that for subsequent screenings, a separate counseling and shared decisionmaking visit (as required for the initial screening) would be overly burdensome. However, we believe that in order to balance the benefits, harms, and complex nature of this service, the beneficiary needs to obtain a written order for subsequent screenings to support continued shared decisionmaking in light of the beneficiary's changing health status, particularly as one ages. For example, the Medicare-covered annual wellness visit (AWV) may be one option for obtaining such an order. The AWV includes elements for the furnishing of appropriate referrals and orders for preventive services.”
“Therefore, we are maintaining the criterion that a beneficiary must receive a written order for subsequent screenings during any appropriate visit with a physician or qualified non-physician practitioner. While we are not requiring a separate counseling and shared decisionmaking visit for subsequent screenings, we are modifying this criterion to note that if a physician or qualified non-physician practitioner chooses to furnish a counseling and shared decisionmaking visit for subsequent screenings, the visit must include all of the components required for an initial counseling and shared decisionmaking visit as defined in this NCD.”
CMS Language About Qualified Providers
Regarding which providers can furnish shared decisionmaking about lung cancer screening, CMS Decision Memo, page 2 of the PDF, states:
“A beneficiary must receive a written order for LDCT lung cancer screening during a lung cancer screening counseling and shared decisionmaking visit, furnished by a physician (as defined in Section 1861(r)(1) of the Social Security Act) or qualified non-physician practitioner (meaning a physician assistant, nurse practitioner, or clinical nurse specialist as defined in §1861(aa)(5) of the Social Security Act).”
“We believe that formal shared decisionmaking between the physician or qualified non-physician practitioner and the beneficiary is an important and essential component to an effective lung cancer screening program. We leave it to the discretion of the physician or qualified non-physician practitioner to determine whether other medical professionals should also participate in the visit based on a particular beneficiary's needs. Therefore, we are maintaining the requirement that the counseling and shared decisionmaking visit must be furnished by a physician or qualified non-physician practitioner.”
CMS Language About Smoking Cessation Counseling
CMS Decision Memo, page 43 of the PDF, states:
“Information regarding Medicare coverage of tobacco cessation counseling services can be found in Pub. 100-03, Medicare National Coverage Determinations Manual, Chapter 1, Sections 210.4 and 210.4.1.”
CMS Language About Auditing/Eligibility of LDCT Facilities
Eligible radiology imaging facilities furnishing lung cancer screening with LDCT are required to submit data to a CMS-approved registry for each lung cancer LDCT screening performed. The list of CMS-approved lung cancer screening registries is available from:
American College of Radiology (ACR) Lung Cancer Screening Registry (LCSR)
Phone: 1-800-227-5463, extension 3535
CMS Decision Memo, page 40 of the PDF, states:
“The primary purpose for requiring the submission of data to the registry is to document compliance with the coverage criteria that are not evidenced on the health care claim.”
CMS Decision Memo, page 4 of the PDF, Radiology imaging facility eligibility criteria, states eligible facility:
Performs LDCT with volumetric CT dose index (CTDIvol) of ≤ 3.0 mGy (milligray) for standard size patients (defined to be 5' 7” and approximately 155 pounds) with appropriate reductions in CTDIvol for smaller patients and appropriate increases in CTDIvol for larger patients;
Utilizes a standardized lung nodule identification, classification and reporting system;
Makes available smoking cessation interventions for current smokers; and
Collects and submits data to a CMS-approved registry for each LDCT lung cancer screening performed. The data collected and submitted to a CMS-approved registry must include, at minimum, all of the following elements:
- Radiologist (reading).
- Ordering Practitioner (National Provider Identifier).
- CT Scanner.
- Smoking history.
- Effective radiation dose.
- Screening information (i.e., Screen date and initial or subsequent screen).
“All CMS-approved registries must have the capacity and capability to collect data from any Medicare-eligible imaging facility/department that furnishes lung cancer screening with LDCT, with a catchment area that includes all 50 States, United States Territories, and the District of Columbia. CMS will evaluate each entity interested in participating as a CMS-approved registry to determine if they are capable of meeting the registry and data collection requirements outlined in this national coverage determination, including:
- Establishment of a steering committee and a governance board for oversight of the registry.
- Registry management plan, including identification of key personnel.
- Operational plan and framework that describes mechanisms for collection and submission of data from imaging facilities to the registry.
- Registry catchment area.
- Mechanisms for the submission of registry data to CMS electronically.
- Mechanisms to collect information (e.g.; HICN) in order to permit linkage of registry data with external databases (e.g., Medicare claims data sets).
- Description of data management and data quality review methods, including validation.
- Use of CMS-approved standardized data dictionary.
- Mechanisms for submitting a list of facilities participating in the registry to CMS.
- Quality assurance plan.
To apply to function as a CMS-approved registry, interested entities must submit a letter of interest along with detailed supporting information about how the interested entity is able to meet the requirements outlined in this national coverage determination.”
Volk: It's an interesting time from a policy perspective, and also from the perspective of clinicians out in practice. This is the first time that we've had policy that endorses shared decisionmaking and specifically calls for patient decisions, so it's a really exciting time for decision scientists to really think about this. But it's also an incredibly challenging time in terms of trying to think about how we're going to make this happen in real-world clinical settings. We're hopeful that the tools that we've developed will be a good starting point. We'd be very excited to hear from you about how you're thinking about using the tools and how you might adapt them for your settings. We hope the toolkit is going to be an invaluable resource going forward, as we're in the midst of this new experience around cancer prevention involving shared decisionmaking.
Page originally created September 2016